Kanoni et al. Genome Biology (2022) 23:268 Genome Biology https://doi.org/10.1186/s13059-022-02837-1 RESEARCH Open Access Implicating genes, pleiotropy, and sexual dimorphism at blood lipid loci through multi‑ancestry meta‑analysis Stavroula Kanoni1†, Sarah E. Graham2†, Yuxuan Wang3†, Ida Surakka2†, Shweta Ramdas4†, Xiang Zhu5,6,7,8†, Shoa L. Clarke7,9, Konain Fatima Bhatti1, Sailaja Vedantam10,11, Thomas W. Winkler12, Adam E. Locke13, Eirini Marouli1, Greg J. M. Zajac14, Kuan‑Han H. Wu15, Ioanna Ntalla16, Qin Hui17,18, Derek Klarin7,19, Austin T. Hilliard7, Zeyuan Wang17,18, Chao Xue2, Gudmar Thorleifsson20, Anna Helgadottir20, Daniel F. Gudbjartsson20,21, Hilma Holm20, Isleifur Olafsson22, Mi Yeong Hwang23, Sohee Han23, Masato Akiyama24,25, Saori Sakaue24,26,27, Chikashi Terao28, Masahiro Kanai11,24,29, Wei Zhou11,15,30, Ben M. Brumpton31,32,33, Humaira Rasheed31,32,34, Aki S. Havulinna35,36, Yogasudha Veturi37, † Stavroula Kanoni, Sarah E. Graham, Yuxuan Wang, Ida Jennifer Allen Pacheco38, Elisabeth A. Rosenthal39, Todd Lingren40, QiPing Feng41, Surakka, Shweta Ramdas, and Xiang Zhu contributed equally to Iftikhar J. Kullo42, Akira Narita43, Jun Takayama43, Hilary C. Martin44, Karen A. Hunt45, this work. Bhavi Trivedi45, Jeffrey Haessler46, Franco Giulianini47, Yuki Bradford37, † Michael Boehnke, Christopher Jason E. Miller37, Archie Campbell48,49, Kuang Lin50, Iona Y. Millwood50,51, D. Brown, Pradeep Natarajan, Panos Deloukas, Cristen J. Willer, Asif Rasheed52, George Hindy53, Jessica D. Faul54, Wei Zhao54,55, David R. Weir54, Themistocles L. Assimes, and Constance Turman56, Hongyan Huang56, Mariaelisa Graff57, Ananyo Choudhury58, Gina M. Peloso jointly supervised this work. Dhriti Sengupta58, Anubha Mahajan59, Michael R. Brown60, Weihua Zhang61,62, *Correspondence: Ketian Yu14, Ellen M. Schmidt14, Anita Pandit14, Stefan Gustafsson63, Xianyong Yin14, cristen@umich.edu; gpeloso@bu Jian’an Luan64, Jing‑Hua Zhao65, Fumihiko Matsuda66, Hye‑Mi Jang23, .edu 2 Kyungheon Yoon23, Carolina Medina‑Gomez67, Achilleas Pitsillides3, Department of Internal Medicine, Division of Cardiology, Jouke Jan Hottenga68,69, Andrew R. Wood70, Yingji Ji70, Zishan Gao71,72,73, University of Michigan, Ann Simon Haworth32,74, Noha A. Yousri75,76, Ruth E. Mitchell32,77, Jin Fang Chai78, Arbor, MI 48109, USA 3 Department of Biostatistics, Mette Aadahl79,80, Anne A. Bjerregaard79, Jie Yao81, Ani Manichaikul82, Boston University School of Public Health, 801 Chii‑Min Hwu83, Yi‑Jen Hung84, Helen R. Warren85,86, Julia Ramirez85,87, Massachusetts Ave, Boston, MA Jette Bork‑Jensen88, Line L. Kårhus79, Anuj Goel59,89, Maria Sabater‑Lleal90,91, 02118, USA Full list of author information is Raymond Noordam92, Pala Mauro93, Floris Matteo93,94, Aaron F. McDaid95,96, available at the end of the article Pedro Marques‑Vidal97, Matthias Wielscher98, Stella Trompet99,100, Naveed Sattar101, Line T. Møllehave79, Matthias Munz102, Lingyao Zeng103,104, Jianfeng Huang105, Bin Yang105, Alaitz Poveda106, Azra Kurbasic106, Claudia Lamina107,108, Lukas Forer107,108, Markus Scholz109,110, Tessel E. Galesloot111, Jonathan P. Bradfield112, Sanni E. Ruotsalainen35, EWarwick Daw113, © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the mate‑ rial. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​ creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​mmons.​org/​publi​ cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Kanoni et al. Genome Biology (2022) 23:268 Page 2 of 42 Joseph M. Zmuda114, Jonathan S. Mitchell115, Christian Fuchsberger115, Henry Christensen116, Jennifer A. Brody117, Miguel Vazquez‑Moreno118, Mary F. Feitosa113, Mary K. Wojczynski113, Zhe Wang119, Michael H. Preuss119, Massimo Mangino120,121, Paraskevi Christofidou120, Niek Verweij122, Jan W. Benjamins122, Jorgen Engmann123,124, Noah L. Tsao125, Anurag Verma4, Roderick C. Slieker126,127, Ken Sin Lo128, Nuno R. Zilhao129, Phuong Le130, Marcus E. Kleber131,132, Graciela E. Delgado131, Shaofeng Huo133, Daisuke D. Ikeda134, Hiroyuki Iha134, Jian Yang135,136,137, Jun Liu138,139, Ayşe Demirkan139,140, Hampton L. Leonard141,142, Jonathan Marten143, Mirjam Frank144, Börge Schmidt144, Laura J. Smyth145, Marisa Cañadas‑Garre145,146,147,148, Chaolong Wang149,150, Masahiro Nakatochi151, Andrew Wong152, Nina Hutri‑Kähönen153, Xueling Sim78, Rui Xia154, Alicia Huerta‑Chagoya155,156,157, Juan Carlos Fernandez‑Lopez158, Valeriya Lyssenko106,159, Suraj S. Nongmaithem161, Swati Bayyana160,161, Heather M. Stringham14, Marguerite R. Irvin162, Christopher Oldmeadow163, Han‑Na Kim164,165, Seungho Ryu166,167, Paul R. H. J. Timmers143,168, Liubov Arbeeva169, Rajkumar Dorajoo150,170, Leslie A. Lange171, Gauri Prasad172,173, Laura Lorés‑Motta174, Marc Pauper174, Jirong Long175, Xiaohui Li81, Elizabeth Theusch176, Fumihiko Takeuchi177, Cassandra N. Spracklen178,179, Anu Loukola35, Sailalitha Bollepalli35, Sophie C. Warner180,181, Ya Xing Wang182, Wen B. Wei183, Teresa Nutile184, Daniela Ruggiero184,185, Yun Ju Sung186, Shufeng Chen105, Fangchao Liu105, Jingyun Yang187,188, Katherine A. Kentistou168, Bernhard Banas189, Giuseppe Giovanni Nardone190, Karina Meidtner191,192, Lawrence F. Bielak55, Jennifer A. Smith55,54, Prashantha Hebbar193, Aliki‑Eleni Farmaki194,195, Edith Hofer196,197, Maoxuan Lin198, Maria Pina Concas190, Simona Vaccargiu199, Peter J. van der Most200, Niina Pitkänen201,202, Brian E. Cade203,204, Sander W. van der Laan205, Kumaraswamy Naidu Chitrala206,207, Stefan Weiss208, Amy R. Bentley209, Ayo P. Doumatey209, Adebowale A. Adeyemo209, Jong Young Lee210, Eva R. B. Petersen211, Aneta A. Nielsen212, Hyeok Sun Choi213, Maria Nethander214,215, Sandra Freitag‑Wolf216, Lorraine Southam44,217, Nigel W. Rayner44,59,217,218, Carol A. Wang219, Shih‑Yi Lin220,221,222, Jun‑Sing Wang223,224, Christian Couture225, Leo‑Pekka Lyytikäinen226,227, Kjell Nikus228,229, Gabriel Cuellar‑Partida230, Henrik Vestergaard88,231, Bertha Hidalgo232, Olga Giannakopoulou1, Qiuyin Cai175, Morgan O. Obura126, Jessica van Setten233, Xiaoyin Li234, Jingjing Liang234, Hua Tang235, Natalie Terzikhan139, Jae Hun Shin213, Rebecca D. Jackson236, Alexander P. Reiner237, Lisa Warsinger Martin238, Zhengming Chen50,51, Liming Li239, Takahisa Kawaguchi66, Joachim Thiery109,240, Joshua C. Bis117, Lenore J. Launer241, Huaixing Li133, Mike A. Nalls242,243, Olli T. Raitakari201,202,244, Sahoko Ichihara245, Sarah H. Wild246, Christopher P. Nelson180,181, Harry Campbell168, Susanne Jäger191,192, Toru Nabika247, Fahd Al‑Mulla193, Harri Niinikoski248,249, Peter S. Braund180,181, Ivana Kolcic250, Peter Kovacs251, Tota Giardoglou252, Tomohiro Katsuya253,254, Dominique de Kleijn255, Gert J. de Borst255, Eung Kweon Kim256, Hieab H. H. Adams139,257,258, M. Arfan Ikram139, Xiaofeng Zhu234, Folkert W. Asselbergs233, Adriaan O. Kraaijeveld233, Joline W. J. Beulens126,259, Kanoni et al. Genome Biology (2022) 23:268 Page 3 of 42 Xiao‑Ou Shu175, Loukianos S. Rallidis260, Oluf Pedersen88, Torben Hansen88, Paul Mitchell261, Alex W. Hewitt262,263, Mika Kähönen264,265, Louis Pérusse225,266, Claude Bouchard267, Anke Tönjes251, Yii‑Der Ida Chen81, Craig E. Pennell219, Trevor A. Mori268, Wolfgang Lieb269, Andre Franke270, Claes Ohlsson214,271, Dan Mellström214,272, Yoon Shin Cho213, Hyejin Lee273, Jian‑Min Yuan274,275, Woon‑Puay Koh276,277, Sang Youl Rhee278, Jeong‑Taek Woo278, Iris M. Heid12, Klaus J. Stark12, Martina E. Zimmermann12, Henry Völzke279, Georg Homuth208, Michele K. Evans280, Alan B. Zonderman280, Ozren Polasek250,281, Gerard Pasterkamp205, Imo E. Hoefer205, Susan Redline203,204, Katja Pahkala201,202,282, Albertine J. Oldehinkel283, Harold Snieder200, Ginevra Biino284, Reinhold Schmidt196, Helena Schmidt285, Stefania Bandinelli286, George Dedoussis194, Thangavel Alphonse Thanaraj193, Sharon L. R. Kardia55, Patricia A. Peyser55, Norihiro Kato177, Matthias B. Schulze191,192,287, Giorgia Girotto190,288, Carsten A. Böger189,289,290, Bettina Jung189,289,290, Peter K. Joshi168, David A. Bennett187,188, Philip L. De Jager11,291, Xiangfeng Lu105, Vasiliki Mamakou292,293, Morris Brown86,294, Mark J. Caulfield85,86, Patricia B. Munroe85,86, Xiuqing Guo81, Marina Ciullo184,185, Jost B. Jonas182,295,296,297, Nilesh J. Samani180,181, Jaakko Kaprio35, Päivi Pajukanta298, Teresa Tusié‑Luna299, Carlos A. Aguilar‑Salinas300, Linda S. Adair301,302, Sonny Augustin Bechayda303,304, H. Janaka de Silva305, Ananda R. Wickremasinghe306, Ronald M. Krauss307, Jer‑Yuarn Wu308, Wei Zheng175, Anneke Iden Hollander174, Dwaipayan Bharadwaj173,309, Adolfo Correa310, James G. Wilson311, Lars Lind312, Chew‑Kiat Heng313, Amanda E. Nelson169,314, Yvonne M. Golightly57,169,315,316, James F. Wilson143,168, Brenda Penninx69,317, Hyung‑Lae Kim318, John Attia163,319, Rodney J. Scott163,319, D. C. Rao320, Donna K. Arnett321, Steven C. Hunt75,322, Mark Walker323, Heikki A. Koistinen36,324,325, Giriraj R. Chandak161,326, Josep M. Mercader156,327,328, Maria C. Costanzo156, Dongkeun Jang156, Noël P. Burtt156, Clicerio Gonzalez Villalpando329, Lorena Orozco330, Myriam Fornage154,331, EShyong Tai78,332, Rob M. van Dam78,333, Terho Lehtimäki226,227, Nish Chaturvedi152, Mitsuhiro Yokota334, Jianjun Liu150, Dermot F. Reilly335, Amy Jayne McKnight145, Frank Kee145, Karl‑Heinz Jöckel144, Mark I. McCarthy59,218, Colin N. A. Palmer336, Veronique Vitart143, Caroline Hayward143, Eleanor Simonsick337, Cornelia M. van Duijn138,139, Zi‑Bing Jin338,339, Jia Qu339, Haretsugu Hishigaki134, Xu Lin133, Winfried März131,340,341, Vilmundur Gudnason129,342, Jean‑Claude Tardif128, Guillaume Lettre128, Leen M.‘t Hart127,343,344,345, Petra J. M. Elders346,347, Scott M. Damrauer4,125,348, Meena Kumari349, Mika Kivimaki124, Pim van der Harst122,233, Tim D. Spector120, Ruth J. F. Loos88,119,350, Michael A. Province113, Esteban J. Parra130, Miguel Cruz118, Bruce M. Psaty117,351,352, Ivan Brandslund116,353, Peter P. Pramstaller115, Charles N. Rotimi209, Kaare Christensen354, Samuli Ripatti35,355,356, Elisabeth Widén35, Hakon Hakonarson357,358, Struan F. A. Grant4,358,359, Lambertus A. L. M. Kiemeney111, Jacqueline de Graaf111, Markus Loeffler109,110, Florian Kronenberg107,108, Dongfeng Gu105,360, Jeanette Erdmann361, Heribert Schunkert103,104, Paul W. Franks106, Allan Linneberg79,80, J. Wouter Jukema99,362, Amit V. Khe Kanoni et al. Genome Biology (2022) 23:268 Page 4 of 42 ra11,363,364,365,366, Minna Männikkö367, Marjo‑Riitta Jarvelin98,368,369, Zoltan Kutalik95,96, Cucca Francesco370,371, Dennis O. Mook‑Kanamori372,373, Ko Willems van Dijk374,375,376, Hugh Watkins89,59, David P. Strachan377, Niels Grarup88, Peter Sever378, Neil Poulter379, Lee‑Ming Chuang380, Jerome I. Rotter81, Thomas M. Dantoft79, Fredrik Karpe381,382, Matt J. Neville381,382, Nicholas J. Timpson32,77, Ching‑Yu Cheng383,384, Tien‑Yin Wong383,384, Chiea Chuen Khor150, Hengtong Li385, Charumathi Sabanayagam383,384, Annette Peters73,104,192, Christian Gieger72,73,192, Andrew T. Hattersley386, Nancy L. Pedersen387, Patrik K. E. Magnusson387, Dorret I. Boomsma68,69, Allegonda H. M. Willemsen68,69, LAdrienne Cupples3,388, Joyce B. J. van Meurs67,139, Mohsen Ghanbari139,389, Penny Gordon‑Larsen301,302, Wei Huang390, Young Jin Kim23, Yasuharu Tabara66, Nicholas J. Wareham64, Claudia Langenberg64,391, Eleftheria Zeggini44,392,217, Johanna Kuusisto393, Markku Laakso393, Erik Ingelsson9,63,394,395, Goncalo Abecasis14,396, John C. Chambers61,62,397,398, Jaspal S. Kooner62,379,399,400, Paul S. de Vries60, Alanna C. Morrison60, Scott Hazelhurst58,401, Michèle Ramsay58, Kari E. North57, Martha Daviglus402, Peter Kraft56,403, Nicholas G. Martin404, John B. Whitfield404, Shahid Abbas52,405, Danish Saleheen52,406,407, Robin G. Walters50,51,408, Michael V. Holmes50,51,409, Corri Black410, Blair H. Smith411, Aris Baras396, Anne E. Justice412, Julie E. Buring47,328, Paul M. Ridker47,328, Daniel I. Chasman47,328, Charles Kooperberg46, Gen Tamiya43, Masayuki Yamamoto43, David A. van Heel45, Richard C. Trembath413, Wei‑Qi Wei414, Gail P. Jarvik415, Bahram Namjou416, M. Geoffrey Hayes417,418,38, Marylyn D. Ritchie37, Pekka Jousilahti36, Veikko Salomaa36, Kristian Hveem31,419,420, Bjørn Olav Åsvold31,419,421, Michiaki Kubo422, Yoichiro Kamatani423,424, Yukinori Okada26,423,425,426, Yoshinori Murakami427, Bong‑Jo Kim23, Unnur Thorsteinsdottir20,428, Kari Stefansson20,438, Jifeng Zhang2, YEugene Chen2, Yuk‑Lam Ho429, Julie A. Lynch430,431, Daniel J. Rader4, Philip S. Tsao7,9,432, Kyong‑Mi Chang433,434, Kelly Cho429,435, Christopher J. O’Donnell429,435, John M. Gaziano429,435, Peter W. F. Wilson18,436, Timothy M. Frayling70, Joel N. Hirschhorn10,11,437, Sekar Kathiresan364,11,365, Karen L. Mohlke178, Yan V. Sun17,18, Andrew P. Morris438, Michael Boehnke14†, Christopher D. Brown4†, Pradeep Natarajan11,439,440,441†, Panos Deloukas1,442†, Cristen J. Willer2,15,443*†, Themistocles L. Assimes7,9,432† and Gina M. Peloso3*†    Kanoni et al. Genome Biology (2022) 23:268 Page 5 of 42 Abstract Background: Genetic variants within nearly 1000 loci are known to contribute to modulation of blood lipid levels. However, the biological pathways underlying these associations are frequently unknown, limiting understanding of these findings and hindering downstream translational efforts such as drug target discovery. Results: To expand our understanding of the underlying biological pathways and mechanisms controlling blood lipid levels, we leverage a large multi-ancestry meta- analysis (N = 1,654,960) of blood lipids to prioritize putative causal genes for 2286 lipid associations using six gene prediction approaches. Using phenome-wide association (PheWAS) scans, we identify relationships of genetically predicted lipid levels to other diseases and conditions. We confirm known pleiotropic associations with cardiovas‑ cular phenotypes and determine novel associations, notably with cholelithiasis risk. We perform sex-stratified GWAS meta-analysis of lipid levels and show that 3–5% of autosomal lipid-associated loci demonstrate sex-biased effects. Finally, we report 21 novel lipid loci identified on the X chromosome. Many of the sex-biased autosomal and X chromosome lipid loci show pleiotropic associations with sex hormones, emphasiz‑ ing the role of hormone regulation in lipid metabolism. Conclusions: Taken together, our findings provide insights into the biological mecha‑ nisms through which associated variants lead to altered lipid levels and potentially cardiovascular disease risk. Keywords: Cholesterol, Lipids, Genetics, Genome-wide association study, GWAS Background Abnormal blood lipid levels are a major cause of cardiovascular disease [1], the lead- ing cause of morbidity and mortality worldwide [2]. Conventional blood lipid measures, low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), and nonHDL-C (TC – HDL-C), are com- monly used in clinical practice to identify individuals at high risk for cardiovascular events. Several treatments for reducing LDL-C, including statins, ezetimibe, and PCSK9 inhibitors [3], also reduce the risk of developing cardiovascular disease. Genome-wide association studies (GWAS) for blood lipids have identified nearly 1000 associated genetic loci to date [4–23], including our recent multi-ancestry GWAS meta-analysis in 1.65 M individuals [24]. The latter focused on the gains from the multi- ancestry meta-analysis relative to the single-ancestry results, in terms of number of loci, fine-mapping, and polygenic score (PGS) transferability. However, a challenge in the field is that the underlying gene and biological pathways is often unknown for GWAS loci. Within lipid GWAS, prior fine-mapping studies combined with functional follow- up have successfully identified causal genes with high confidence for only a handful of associated GWAS loci, including SORT1 [25], TM6SF2 [12], and ANGPTL3 [26], among others. Highly sophisticated methods are emerging to prioritize causal genes in well- powered GWAS studies, such as the Data-driven Expression-Prioritized Integration for Complex Traits [27] (DEPICT) and the Polygenic Priority Score [28] (PoPS), that take into account genome-wide properties of associated loci and larger sets of associ- ated loci are beneficial. These methods can be combined with algorithms that integrate expression data such as transcriptome-wide association studies (TWAS) and compre- hensive experimental data sets such as mouse gene knockouts. Gene sets enriched for Kanoni et al. Genome Biology (2022) 23:268 Page 6 of 42 causal genes will enhance our ability to unravel the biological pathways underlying these associations and there is growing interest in using a combination of gene prioritization methods to provide compelling evidence for putative causal genes [29]. In parallel, the linkage of electronic health records with genetic data in large-scale population studies and patient biobanks allows for the systematic exploration of pleiot- ropy of lipid-associated alleles. While blood lipid levels have a well-documented causal effect on cardiovascular disease based on genetic association studies validated by rand- omized controlled trials [30–32], genetic pleiotropic associations might exist for other conditions. Unraveling such pleiotropy may yield new biological insights by revealing previously unrecognized connections between blood lipids and both cardiovascular and non-cardiovascular diseases. Phenome-wide association scans (PheWAS) adopt an agnostic approach to test for pleiotropic associations between genetic factors and a wide range of phenotypes [33]. Such knowledge may allow for the identification of lipid levels as novel diagnostic biomarkers, the repurposing of drugs, and the prevention of adverse drug events [34]. Finally, given empirical sex differences in blood lipid distributions, sex-specific genetic associations may yield novel biological insights. Pre-menopausal females have lower lev- els of LDL-C than same-age males, and HDL-C levels are higher among females of all ages compared to males [35]. Lipid levels also show a greater estimated heritability in females compared with males [36], especially for LDL-C and TC (> 1.3-fold difference). Sexual dimorphism in lipid levels may be partly explained by X chromosome variants. Evidence from human X-linked abnormalities (like Turner or Klinefelter syndromes) suggests an important role of this chromosome in lipid metabolism [37]. This is further corroborated by the lipid and atherosclerosis profiles in the Four Core Genotypes mouse model [38], which comprises XX and XY gonadal males and XX and XY gonadal females. GWAS studies have traditionally understudied the X chromosome due to technical and analytical difficulties. A recent high coverage whole X chromosome sequencing study [39] prioritized CHRDL1 as a candidate causal lipid gene, suggesting with larger sample sizes we may be able to discover additional variation on the X chromosome associated with lipid levels. In this study, we first prioritize genes at GWAS lipid loci through multiple in silico gene prediction algorithms and experimental data sources using the latest Global Lipids Genetics Consortium multi-ancestry meta-analysis [24]. We then identify novel disease associations related to lipid levels through PheWAS in two large biobanks using PGSs. Finally, we perform sex-stratified meta-analysis to compare the associations between males and females to identify genetic loci with sex-specific associations and GWAS meta-analysis of the X chromosome, to better understand lipid level differences between the sexes. Together, our results highlight biological mechanisms through which lipid- associated variants lead to altered lipid levels. Results Identifying functional genes in lipid‑associated loci In a GWAS meta-analysis of blood lipid levels from 1.65 million individuals (Additional file 1: Table S1) at 91 million genotyped and imputed genetic variants, we observed a total of 2286 genome-wide significant index variants associated with lipid levels at 923 Kanoni et al. Genome Biology (2022) 23:268 Page 7 of 42 loci (± 500 kb regions). This corresponded to 416 index variants associated with LDL- C, 539 with HDL-C, 461 with TG, 487 with TC, and 383 with nonHDL-C. Uniquely, we observed 1750 variants associated with one or more lipid levels [24] (Additional file 2: Table S2). We employed six approaches to identify candidate functional genes for all 2286 lipid associations. Our prioritization approaches include four locus-specific methods that are based on local information around the indexed variant: (1) the closest gene to the index variant, (2) genes with lipid-associated protein-altering variants, (3) colocalized expression quantitative trait loci (eQTL) genes, and (4) nearby genes prioritized by tran- script-wide association studies (TWAS). We also used two genome-wide methods that leveraged genome-wide similarities of features: (1) gene-level Polygenic Priority Score (PoPS) [28] and (2) Data-driven Expression-Prioritized Integration for Complex Traits (DEPICT) [27]. We further combined the two genome-wide methods with the locus- specific methods to increase the confidence in prioritized genes: (1) PoPS intersects with any locus-specific methods (PoPS +), and (2) DEPICT intersects with any locus-specific methods (DEPICT +) (Fig. 1). Since the genome-wide gene prioritization approaches can prioritize different genes for different lipid types at the same locus, we report the gene prioritization results for all 2286 lipid-variant associations (Additional file 2: Table S2, Additional file 3: Figure S1). We took the genes prioritized by PoPS + and performed text mining to determine whether previous biological evidence supported these genes as playing a role in lipid lev- els (Additional file 4: Table S3, S4). PoPS + leverages both locus-specific and genome- wide genetic signals to increase confidence level in prioritized genes [28]. In total, 882 out of 2286 lipid associations were assigned to one potential causal gene based on PoPS + . We identified a group of 466 unique genes among the 882 lipid associations. We determined that 31 out of the 466 PoPS + genes have over 1000 lipid-related pub- lications, 91 PoPS + genes have 100–999 lipid-related publications, 321 PoPS + genes have 1–99 lipid-related publications, and 23 PoPS + genes had no lipid-related publica- tions retrieved by the text mining algorithm. These 23 genes could indicate novel genes related to lipid levels for future work or be due to incorrect gene prioritization for a small fraction of index variants. (Additional file 5: Table S4). We then randomly selected 466 genes from 18,383 protein-coding genes using by the PoPS as the reference group to estimate the number of lipid-related publications we would expect to see by chance. A Mann–Whitney U test showed that there was a significant difference (W = 52,353, p-value < 2.2 x ­10−16) between the set of genes identified by PoPS + compared to the ref- erence set of 466 genes (Additional file 6: Figure S2). The median count of lipid-related publications was 19 for the PoPS + gene set compared with 2 lipid-related publications for genes in the reference set. We performed a comprehensive lookup of all PoPS + prioritized lipid genes in the Therapeutic Target Database 2022 [24] and found 2092 drugs targeting at least one of our 102 PoPS + prioritized lipid genes observed in the database (Additional file 7: Table S5). Among those 102 PoPS + genes, we identify known drug target genes including PCSK9 druggable as subtilisin/kexin type 9 inhibitor, HMGCR​ druggable as HMG-CoA reduc- tase inhibitor, PDE3A druggable as phosphodiesterase 3A inhibitor (CILOSTAZOL), and NR1H4 as a bile acid receptor FXR agonist (URSODIOL). We also identify several Kanoni et al. Genome Biology (2022) 23:268 Page 8 of 42 Fig. 1 Schematic of multi-method candidate gene mapping of indexed variants associated with blood lipid levels. We defined indexed variants within the GLGC GWAS summary statistics and performed two similarity-based methods and four locus-based methods to prioritize genes for each of the indexed variants other potential drug targets [24] such as LIPG (lipase G) and NR1H3 (nuclear receptor subfamily 1 group H member 3), with relevant lipid biology. LIPG has phospholipase and triglyceride lipase activities and is a primary determinant of plasma HDL levels. NR1H3 has an important role in the regulation of cholesterol homeostasis, regulating cholesterol uptake through MYLIP-dependent ubiquitination of LDLR, VLDLR, and LRP8 that could be targeted as an LXR-alpha modulator. Effects of protein‑altering lipid alleles with protective effects on CAD, T2D, and NAFLD Coronary artery disease (CAD), type 2 diabetes (T2D), and non-alcoholic fatty liver dis- ease (NAFLD) are typically characterized by dyslipidemias. We examined protein-alter- ing alleles with favorable lipid profiles for their associations with CAD, T2D, and NAFLD to identify potential cardiovascular drug targets without off-target liver or diabetes effects. Of the 2286 lipid associations, we observed 166 coding index variants. Eight- een coding variants with a protective lipid effect also had a protective effect for CAD Kanoni et al. Genome Biology (2022) 23:268 Page 9 of 42 or T2D (p-value < 0.001) and the lipid results colocalized with the CAD or T2D results, as appropriate, with a posterior probability of a shared causal variant > 0.8 (Table 1 and Additional file 8: Table S6). Six of these twenty variants had protective effects for both CAD and T2D, while nine were protective for CAD and three were protective for T2D (Table 1). Additionally, 269 noncoding alleles with a protective lipid effect also had a protective effect for CAD or T2D (p < 0.001; Additional file 8: Table S6). Driver tissues for lipid levels We applied DESE (Driver-tissue Estimation by Selective Expression) [40] to estimate the driver tissues of five lipid traits using both gene-level and transcript-level eQTL sum- mary statistics from GTEx v8 tissues [41]. We identified liver as the top-ranked tissue for HDL-C (gene-level p-value = 4.5 x ­10−18, transcript-level p-value = 3.0 x ­10−26), TC (gene-level p-value = 1.1 x ­10−25, transcript-level p-value = 1.4 x ­10−33), and nonHDL- C (gene-level p-value = 2.0 x ­10−19, transcript-level p-value = 3.9 x ­10−29) based on both gene-level and transcript-level selective expression (Additional file 9: Figure S3, Additional file 10: Table S7). For LDL-C, we identified the spleen as the top-ranked tis- sue using GTEx gene-level data (p-value = 8.3 x ­10−20), while liver was ranked second (p-value = 4.8 x ­10−17). However, when using GTEx transcript-level data, liver was the top-ranked tissue (p-value = 4.3 x ­10−29) and second was whole blood (p-value = 4.3 x ­10−20). The top tissue for TG according to both GTEx gene-level and transcript-level expression data was whole blood (gene-level p-value = 6.4 x ­10−20, transcript-level p-value = 1.4 x ­10−21). Spleen and liver were second according to GTEx gene-level and transcript-level expression data, respectively. The results were consistent with previous knowledge that the liver is a major tissue for lipid metabolism. Transcript-level selective expression provided more statistically significant results for the estimated driver tissues compared to the gene-level selective expression, as reported in the original [40]. Polygenic scores for lipid phenotypes and phenome‑wide association scans We have previously reported that a polygenic score (PGS) for LDL-C was most informa- tive when generated from the multi-ancestry GWAS and that the multi-ancestry PGS performed equally well in European-ancestry Americans, African-ancestry Americans, and continental Africans [24]. Using a similar approach, we generated PGS for the other four lipid traits (“Methods”). We next performed a phenome-wide association scan (PheWAS) for the multi-ances- try lipid PGS (LDL-C PGS previously reported [24]) to identify pleiotropic effects of lipids with other traits in the European subsets of the UK Biobank and the Million Vet- eran Program (MVP) cohorts. We compared the effect sizes from the PheWAS analysis between the UK Biobank and MVP per lipid PGS and observed a moderate correlation between the two datasets (Additional file 11: Figure S4). The correlation of the PGS effects on all phenotypes between the UK Biobank and MVP ranges from 0.12 for the HDL-C PGS to 0.39 for the TC PGS (Additional file 11: Figure S4). In general, correla- tions were stronger for the ICD-10-based phecodes (r2 of 0.42–0.52) compared to the biomarkers (r2 of 0.06–0.23) (Additional file 11: Figure S4), which may reflect differences in study populations due to varied environmental effects, prevalence of chronic health conditions, and sex distribution. Among PheWAS results with p-value ≤ 0.05 in the UK Table 1 Protective lipid coding alleles associated with CAD and/or T2D RSID Trait Coding variant Effect allele (EAF) Lipid effect CAD OR CAD T2D OR T2D NAFLD OR NAFLD (95% CI) P-value (95% CI) P-value (95% CI) P-value Kanoni et al. Genome Biology Protective lipid alleles associated with reduced risk of CAD and T2D rs116843064 TG ANGPTL4 A (0.020) − 0.238 0.87 1.92x10−18 0.91 2.30x10−5 0.99 0.85 p.Glu40Lys (0.85,0.90) (0.86,0.95) (0.84,1.15) rs1169288 TC HNF1A A (0.682) − 0.035 0.97 2.31x10−16 0.95 7.30x10−13 0.95 0.04 p.Ile27Leu (0.96,0.98) (0.94,0.96) (0.91,1.00) (2022) 23:268 rs2307111 HDL-C POC5 C (0.440) 0.016 0.99 9.09x10−05 0.95 3.30x10−16 1.00 0.85 p.His36Arg (0.98,0.99) (0.94,0.96) (0.96,1.05) rs6480771 HDL-C DUSP13 T (0.531) 0.008 0.99 1.44x10−04 0.97 4.40x10−05 0.92 5.21x10−04 p.Ser111Gly (0.98,0.99) (0.96,0.99) (0.88,0.97) rs35742417 TG RREB1 A (0.173) − 0.012 0.98 5.06x10 −04 0.96 3.70 ­x10−06 0.98 0.60 p.Ser1554Tyr (0.97,0.99) (0.95,0.98) (0.93,1.04) rs72681869 TG SOS2 C (0.008) − 0.053 0.93 3.71x10−03 0.88 3.90x10−04 0.87 0.29 p.Pro191Arg (0.89,0.98) (0.82,0.94) (0.68,1.12) Protective lipid alleles associated with reduced risk of CAD rs7412 LDL-C APOE T (0.076) − 0.517 0.90 9.94x10−52 1.01 0.55 1.01 0.84 p.Arg176Cys (0.88,0.91) (0.98,1.03) (0.93,1.10) rs11591147 LDL-C PCSK9 T (0.015) − 0.434 0.80 5.97x10−36 1.04 0.16 1.05 0.58 p.Arg46Leu (0.77,0.83) (0.99,1.09) (0.88,1.26) rs11601507 LDL-C TRIM5 C (0.926) − 0.042 0.95 2.80x10−12 0.99 0.26 1.02 0.72 p.Val112Phe (0.94,0.96) (0.96,1.01) (0.93,1.11) rs1132274 HDL-C RRBP1 C (0.827) 0.017 0.97 3.57x10−08 1.01 0.43 1.00 0.91 p.Arg891Leu (0.96,0.98) (0.99,1.02) (0.94,1.07) rs4760 HDL-C PLAUR​ A (0.860) 0.016 0.97 7.31x10−07 0.99 0.3 0.96 0.26 p.Leu317Pro (0.96,0.98) (0.97,1.01) (0.91,1.03) rs855791 LDL-C TMPRSS6 G (0.578) − 0.009 0.98 1.08x10−06 1.00 0.75 0.94 4.83x10−03 p.Val736Ala (0.97,0.99) (0.99,1.01) (0.9,0.98) rs58542926 TC TM6SF2 T (0.073) − 0.124 0.97 4.02x10−06 1.10 2.00x10−14 1.45 1.05x10−16 p.Glu167Lys (0.95,0.98) (1.07,1.12) (1.33,1.58) rs56196860 HDL-C FKBP4 A (0.027) 0.031 0.95 1.05x10-05 0.98 0.33 1.03 0.73 p.Asn197Lys (0.92,0.97) (0.94,1.02) (0.87,1.21) Page 10 of 42 Kanoni et al. Genome Biology (2022) 23:268 Table 1 (continued) RSID Trait Coding variant Effect allele (EAF) Lipid effect CAD OR CAD T2D OR T2D NAFLD OR NAFLD (95% CI) P-value (95% CI) P-value (95% CI) P-value rs72836561 HDL-C CD300LG C (0.971) 0.187 0.95 (0.93,0.98) 1.34x10−04 0.98 (0.95,1.02) 0.4 0.98 (0.86,1.12) 0.77 p.Arg82Cys Protective lipid alleles associated with reduced risk of T2D rs1800961 HDL-C HNF4A C (0.969) 0.134 0.99 0.3751 0.85 3.20x10− 20 1.02 0.74 p.Thr139Ile (0.97,1.01) (0.82,0.88) (0.90,1.17) rs1801253 TG ADRB1 C (0.732) − 0.011 1.01 8.76x10− 03 0.97 1.9010x− 05 1.00 0.85 p.Gly389Arg (1.00,1.02) (0.96,0.98) (0.95,1.06) rs13107325 HDL-C SLC39A8 C (0.941) 0.082 1.00 0.8876 0.95 3.30Ex10− 04 0.85 1.85x10− 04 p.Ala391Thr (0.98,1.01) (0.93,0.98) (0.78,0.93) Page 11 of 42 Kanoni et al. Genome Biology (2022) 23:268 Page 12 of 42 Biobank, the correlation was even higher for ICD-10-based phecodes (r2 of 0.52–0.76) but remained the same for the biomarkers (r2 of 0.07–0.22). We meta-analyzed the results from the two cohorts to increase the power of the PheWAS, by matching ICD10-mapped phecodes and biomarkers. In the combined the UK Biobank-MVP PheWAS results, we detected 58 phenotypes associated with the LDL-C PGS at phenome-wide significance level (p-value ≤ 6.5 × ­10−5, corrected for 773 phenotypes), 165 with the HDL-C PGS, 59 with the TC PGS, 166 with the TG PGS, and 78 with the nonHDL-C PGS (Fig. 2, Additional file 12: Table S8, Additional file 13: Figure S5, Additional file 14: Figure S6, Additional file 15: Figure S7, Additional file 16: Figure S8). As expected, multiple cardiovascular phenotypes related to atherosclerosis, including the expected coronary artery disease as well as aortic aneurysm and essen- tial hypertension, were phenome-wide significantly associated with all five lipid PGSs, indicating increased risk of these diseases for individuals with genetically predicted increased LDL-C, TG, TC, or nonHDL-C or genetically predicted decreased HDL-C. A recent wide-ranging Mendelian randomization analysis confirmed the causal effect of circulating lipids, not only for coronary artery disease, but other cardiovascular out- comes [42]. Additionally, all lipid PGSs were also significantly associated with decreased Fig. 2 PheWAS meta-analysis results for multi-ancestry LDL-C PGS in the UK Biobank and MVP. The blue horizontal line denotes phenome-wide significance (p-value ≤ 6.5 × ­10−5, to account for multiple testing of 773 phenotypes) and the red line is genome-wide significance (p-value ≤ 5 × ­10−8). Phenotypes have been pruned, so that the most significant one per correlated phenotype group (correlation coefficient > 0.2) is retained. Pairwise correlations were estimated with chi-square test and Cramer’s V for the dichotomous phenotypes and Pearson’s correlation for the continuous phenotypes. AAA: abdominal aortic aneurysm, AD: Alzheimer’s disease, AST: aspartate aminotransferase, Atherosclerosis*: atherosclerosis of native arteries of the extremities with intermittent claudication, Hb_con: hemoglobin concentration, IBD: irritable bowel disease, MCH: mean corpuscular hemoglobin, MCV: mean corpuscular volume, PVD: peripheral vascular disease Kanoni et al. Genome Biology (2022) 23:268 Page 13 of 42 levels of direct bilirubin (Additional file 12: Table S8, Fig. 2, Additional file 13: Figure S5, Additional file 14: Figure S6, Additional file 15: Figure S7, Additional file 16: Figure S8), indicating genetically predicted lower LDL-C increased levels of bilirubin (Fig. 2). Cor- respondingly, lipid PGSs were associated with lower risk for cholelithiasis (gallstones) with the opposite direction for TG PGS, indicating that extreme lowering of LDL-C may impact rates of cholelithiasis (Additional file 12: Table S8, Fig. 2, Additional file 13: Fig- ure S5, Additional file 14: Figure S6, Additional file 15: Figure S7, Additional file 16: Fig- ure S8). To further clarify whether this association might be driven by the ABCG8 gene alone, we excluded from the LDL-PGS all variants within the locus and tested the associ- ation between LDL-PGS and cholelithiasis in the UK Biobank. There was no attenuation of the observed association (OR = 0.94 and p-value = 7.94 × ­10−17 without the ABCG8 locus vs. OR = 0.93 and p-value = 1.96 × ­10−21). In the PheWAS analysis, we found that the TC and LDL-C PGS were signifi- cantly associated with increased levels of HbA1c (beta = 0.101 mmol/mol per SD PGS increase, p -value= 1.21 × ­10−23 and beta = 0.095 mmol/mol per SD PGS increase, p-value = 4.37 × ­10−21, respectively), while the HDL-C PGS was associ- ated with decreased levels of HbA1c (beta =  − 0.257 mmol/mol per SD PGS increase, p-value = 2.84 × ­10−143) (Additional file 12: Table S8). Furthermore, genetically predicted increased LDL-C was significantly associated with decreased hemoglobin concentration (p-value = 1.92 × ­10−45, similar significant associations for all other lipid PGSs with a reverse direction of effect for TG, Additional file 12: Table S8). As expected, genetically predicted increased LDL-C and TC were both associated with increased risk for Alzhei- mer’s disease [43] (OR = 1.33 per SD PGS increase, p-value = 1.74 × ­10−44 and OR = 1.26 per SD PGS increase, p-value = 1.48 × ­10−30, respectively; Additional file 12: Table S8). To further investigate how this association might be driven by the ApoE locus, we excluded all genetic variants overlapping this gene from the LDL-PGS and repeated the analysis in the UK Biobank. While the association between the LDL-PGS and the risk for Alzheimer’s disease was slightly attenuated after removing the ApoE locus (OR = 1.23 vs. 1.36 per SD PGS increase), the association remained significant (p-value = 2.51 × ­10−21). Recent Mendelian randomization studies also provide evidence for the causal effect of lipids on risk for dementia [44] and Alzheimer’s disease [45]. The LDL-C and TC PGSs were also associated with increased aspartate aminotransferase levels (a liver enzyme), in accordance with other studies [46]. We also observed inverse associations between LDL- PGS (p-value = 1.43 × ­10−14) and TC PGS (p-value = 8.34 × ­10−14) with the risk of iron metabolism disorders (Additional file 12: Table S8). To better understand the loci that drive the association between each of the lipid PGSs and cholelithiasis and cholecystitis, we interrogated the results from the single-variant PheWAS meta-analysis in the UK Biobank and MVP with all lipid multi-ancestry index variants (N = 1750 unique). We identified 22 genetic variants associated with chole- lithiasis and/or cholecystitis at genome-wide significance. Genes prioritized for these index variants included genes already reported to be associated with gallstone disease [47] (CYP7A1, ABCG5/8), as well as additional genes (ABCB4, LRBA, HNF4A, NUCB1, GATA4), that may play also a role. Importantly, we found there was overlap (same index variant) between the previously published index variants for gallstone disease and our lipid index variants for these two loci (Additional file 17: Table S9). Kanoni et al. Genome Biology (2022) 23:268 Page 14 of 42 Lipid loci show sex‑specific effects Sex-stratified analyses have the potential to identify loci missed by sex-combined analy- ses [48] as well as to detect loci exhibiting differential effects on lipids between sexes. First, we performed GWAS meta-analysis separately in each sex (Nmales = 749,391; Nfe- males = 562,410), excluding loci discovered in the sex-combined analysis [24]. We iden- tified twelve loci in females and four in males that reached genome-wide significance in the sex-stratified analysis (p-value < 5 × ­10−8; Additional file 18: Table S10, Additional file 19: Table S11, Additional file 20: Table S12) but not in the sex-combined meta-anal- ysis. As variants may show association to a single sex for reasons unrelated to biological sex differences, including differences in sample sizes between groups, we additionally tested for heterogeneity by sex for these variants in GLGC participating cohorts with close to equal number of males and females. Of the sixteen loci, eight showed significant sex-heterogeneity (p-value < 0.0031, Bonferroni-corrected threshold for sixteen tests). For example, the non-synonymous variant rs34372369 (EPHA1, p.Pro582Leu) is asso- ciated with nonHDL-C only in females (male p-value > 0.05) and shows significant sex- heterogeneity (p-value = 0.0012). This variant has been previously found to be linked with expression levels of the sex hormone-binding globulin gene (SHBG) more strongly in males than females [49], suggesting a possible reason for the difference in observed associations. We additionally sought to replicate the sex-heterogeneity results of these variants in 8 independent multi-ancestry cohorts (N = 311,639, 77% non-European ancestry, Additional file 21: Table S13). However, we did not detect significant differ- ences in effect sizes between sexes for these variants after accounting for the number of tests (p-value > 0.0031, Additional file 22: Table S14), potentially due to the limited sam- ple size or difference in ancestry makeup. Second, we tested for a difference in the male- and female-specific effect sizes for each of the index variants identified from the sex-combined multi-ancestry meta-analysis. Of the 1750 unique index variants, 64 showed a significant difference in effect size by sex for one or more traits (Bonferroni correction for the number of index variants in each trait, Additional file 23: Table S15). These were evenly distributed across traits and more often had stronger effects in females than males (67%, Additional file 24: Figure S9). We tested for replication of the sex-specific differences in up to 311,120 participants from eight independent multi-ancestry cohorts not included in the original meta-analysis (Addi- tional file 21: Table S13). Fifty-four of the 64 (84%) variants had effect size differences that were directionally consistent with the original meta-analysis (Additional file 25: Table S16). Of these, 10 had significantly different effect sizes (p-value < 7.8 × ­10−4, Bon- ferroni correction for 64 variants) and 22 were nominally significant (p-value < 0.05). We attribute the low rate of replication to the small sample size and the differing proportions of ancestry groups within our replication samples, but we cannot dismiss the potential of false positives in the sex-specific discovery results. We tested whether the observed sex differences could be caused by a higher frequency of cholesterol-lowering medications in males, potentially indicating an insufficient cor- rection for pre-medication cholesterol levels. Among white British individuals in the UK Biobank, variants with significant sex differences had significantly higher effect size estimates on average after excluding individuals on medication (Additional file 26: Figure S10, Additional file 27: Table S17). However, of the 17 variants that exhibited a Kanoni et al. Genome Biology (2022) 23:268 Page 15 of 42 significant difference in effect size by sex in the UK Biobank alone, 15 remained sig- nificant after excluding individuals taking medications. Based on this observation, the observed differences did not appear to be driven solely, or even primarily, by differences in medication use between sexes. Furthermore, none of the identified sex-specific vari- ants were associated with sex-participation bias [50] (Additional file 28: Table S18), indi- cating that differential study enrollment between sexes was unlikely to be the cause of the observed sex-specific lipid associations. We next investigated differences in environ- mental factors between sexes for these variants in the UK Biobank (Additional file 29: Table S19), including alcohol use [48], smoking status [48], body mass index (BMI) [51], and waist-hip ratio adjusted for BMI (WHRadjBMI) [51]. Twenty-two of the variants (34%) with differential effects on lipids by sex also exhibited a significant difference by sex for WHRadjBMI and one variant had a significant difference by sex for alcohol use (ADH1B p.His48Arg). The observed sex differences may therefore be partially attributed to pleiotropic associations with other traits. Finally, we annotated each locus that showed significant sex differences with regula- tory information to identify biological mechanisms that could underlie this difference. Of the 64 lipid variants with significant sex-stratified associations, 14 colocalized (pos- terior probability of H4 > 0.8) with expression of 20 genes in lipid-related tissues (liver, adipose, or skeletal muscle; Additional file 30: Table S20). Eight of these loci also show a sex-biased eQTL effect in at least one tissue in the direction concordant with the observed sex specificity of the GWAS effect (Additional file 30: Table S20). Among these ten is CETP, a gene with strong prior evidence for association with lipids, and UGT2B17 [20] (Additional file 31: Supplementary Note, Fig. 3). The lead variant of UGT2B17, rs4860987, shows a significantly stronger effect of LDL-C in males (­Betamale = 0.042, ­SEmale = 0.002, ­Betafemale = 0.016, ­SEfemale = 0.003, p-valuedifference = 4.2 × ­10−15) and colocalizes with a male-specific liver eQTL associated with increased expression of UGT2B17. Common variants at this locus are in moderate LD (R2 = 0.51) with a com- mon copy number variation (CNV), which may mediate the causal effect (Additional file 31: Supplementary Note). UGT2B17 plays a role in the metabolism of androgens [52], including testosterone, which is consistent with the observed pleiotropic relation- ship of this locus with testosterone in males (Additional file 30: Table S20). We note that the index variant in UGT2B17, rs4860987, did not show significant sex-specific effects in the replication cohorts, but this could be due to varying frequencies for the index vari- ant between ancestry groups and the moderate LD to the causal CNV in the region. We observed that the combined frequency of rs4860987 across the replication studies was much lower (8%) compared with our combined frequency in the discovery (24%) due to differing proportions of ancestry groups and, along with the lower number of individuals (N = 218,437), led to a much-reduced power to replicate this sex-specific effect. Lipid‑associated loci on the X chromosome Lastly, we meta-analyzed association statistics for 3.1 million X chromosomal variants, including PAR regions, across 1,238,180 individuals from multiple ancestry groups. We identified 28 index variants significantly associated with lipid levels (Additional file 32: Table S21), of which 21 have not been previously reported [20, 39, 53] (15 for HDL-C, 4 for LDL-C, 4 for TC, 5 for TG and 4 for nonHDL-C, Table 2). Among these 28 loci, two Kanoni et al. Genome Biology (2022) 23:268 Page 16 of 42 Fig. 3 Sex specificity at the UGT2B17 Locus. A The association signal for LDL-C (top panel) colocalizes with the UGT2B17 eQTL signal in the liver (bottom panel). B The effect sizes of this variant on LDL-C and UGT2B17 expression are both significantly higher for males (red) compared to females (blue) Table 2 Novel X chromosome lipid-associated loci RSID Position in EA/NEA Annotation Associated trait EAF N Effect size Ancestry and GC Sex Kanoni et al. Genome Biology chromosome X (hg19) (closest gene) (SE) corrected p-value difference from METAL from MR-MEGA p-value rs35143646 2,856,155 T/C Missense (ARSL) LDL-C 0.6352 1,038,070 0.0115 1.83x10− 16 0.1791 (0.0014) TC 0.6449 1,100,310 0.0091 1.47x10− 14 0.08484 (0.0014) (2022) 23:268 nonHDL-C 0.6822 712,983 0.0112 2.08x10− 11 0.5294 (0.0017) rs5934507 8,917,206 G/A Intergenic (FAM9B) HDL-C 0.274 1,158,000 − 0.0076 (0.0012) 9.99x10− 10 5.83x10− 4 rs191084933 14,133,208 G/A Intergenic (GEMIN8) HDL-C 0.0055 1,052,630 0.0694 5.20x10− 10 0.3272 (0.0101) rs7888119 16,813,128 T/C Intronic (TXLNG) HDL-C 0.546 1,158,000 0.0114 1.47x10− 20 0.2212 (0.0012) rs2230488 20,204,461 T/G Synonymous (RPS6KA3) LDL-C 0.1772 1,135,110 − 0.0143 (0.0015) 6.90x10− 18 0.9297 TC 0.1745 1,237,380 − 0.0114 (0.0014) 4.68x10− 12 0.8906 rs6527977 20,322,238 A/C Intergenic (RPS6KA3) nonHDL-C 0.1602 789,200 − 0.0147 (0.0020) 3.06x10− 10 0.7000 rs12012576 21,813,178 G/A Intergenic (SMPX) TG 0.3103 1,160,340 − 0.0072 (0.0012) 4.88x10− 8 0.4558 rs6609434 46,636,767 C/A Intergenic (SLC9A7) HDL-C 0.4959 767,051 0.0060 4.25x10− 8 0.9035 (0.0014) rs113957181 49,848,600 A/C Intronic (CLCN5) HDL-C 0.0566 452,268 − 0.0297 (0.0047) 1.39x10− 9 0.1896 rs782397956 53,993,589 T/C Intronic (PHF8) HDL-C 0.2050 621,342 − 0.0009 (0.0021) 2.65x10− 15 0.2585 rs72305711 55,981,911 T/TTA​ Intronic (RP13-188A5.1) HDL-C 0.1111 534,967 − 0.0167 (0.0033) 2.22x10− 22 0.03202 rs5914559 56,139,739 G/T Intergenic (KLF8) TG 0.6697 738,155 − 0.0131 (0.0016) 4.44x10− 16 0.3224 rs5964416 64,368,487 A/C Intergenic (ZC4H2) HDL-C 0.0924 614,283 − 0.0316 (0.0016) 2.68x10− 11 0.1478 rs5965342 66,204,144 T/C Intergenic (EDA2R) HDL-C 0.2037 745,721 − 0.0212 (0.0020) 1.20x10− 29 0.02050 rs505520 66,258,914 C/A Intergenic (EDA2R) TG 0.2191 750,415 0.0217 2.91x10− 30 0.004371 (0.0019) rs771540123 67,967,645 A/G Intergenic (STARD8) nonHDL-C 0.0027 15,311 − 0.0202 1.61x10− 9 NA (0.115) rs5937000 70,047,788 C/T Intronic (TEX11) HDL-C 0.4515 796,971 0.0075 9.55x10− 9 0.02603 (0.0013) Page 17 of 42 Kanoni et al. Genome Biology Table 2 (continued) RSID Position in EA/NEA Annotation Associated trait EAF N Effect size Ancestry and GC Sex chromosome X (hg19) (closest gene) (SE) corrected p-value difference (2022) 23:268 from METAL from MR-MEGA p-value rs5938008 74,496,225 C/T Intronic (UPRT) HDL-C 0.9167 1,072,550 0.0174 1.24x10− 8 0.7235 (0.003) rs1802288 99,890,204 T/C Missense (TSPAN6) LDL-C 0.1665 714,113 0.0159 1.42x10− 16 0.9642 (0.0019) TC 0.1684 753,479 0.0128 2.49x10− 10 0.8646 (0.0018) rs139144471 117,829,694 G/T Intergenic (DOCK11) HDL-C 0.0876 1,098,540 − 0.0121 (0.0022) 1.16x10− 8 0.1111 rs6648533 122,804,678 C/T Intronic (THOC2) HDL-C 0.299 1,106,090 − 0.0100 (0.0013) 2.24x10− 12 0.4516 rs5929738 135,265,287 C/A Intronic (FHL1) HDL-C 0.4679 970,330 0.0071 3.21x10− 10 0.4197 (0.0011) rs5975692 135,266,089 G/A Intronic (FHL1) TG 0.465 1,138,740 − 0.0084 (0.0011) 7.87x10− 12 0.9215 rs2070826 153,582,198 C/T Intronic (FLNA) HDL-C 0.1427 1,141,020 0.0170 6.65x10− 26 0.08671 (0.0017) rs11593 153,627,145 C/A Intronic (RPL10) TG 0.1586 1,143,360 − 0.0166 (0.0017) 5.76x10− 21 0.8695 rs7886627 153,679,609 G/A Intergenic (FAM50A) nonHDL-C 0.1214 771,706 − 0.0147 (0.0022) 9.26x10− 9 0.06819 rs1050828 153,764,217 T/C Missense (G6PD) LDL-C 0.0113 744,968 − 0.0514 (0.0061) 2.54x10− 15 0.6533 rs762517 153,764,734 A/G Intronic (G6PD) TC 0.0142 798,600 − 0.0480 (0.0057) 5.26x10− 16 0.7594 EA Effect allele, NEA Non-effect allele, EAF Effect allele frequency, N Number of samples, SE Standard error of the effect size, GC Genomic control, LDL-C Low-density lipoprotein cholesterol, TC Total cholesterol, HDL-C High-density lipoprotein cholesterol, TG Triglycerides Page 18 of 42 Kanoni et al. Genome Biology (2022) 23:268 Page 19 of 42 have index variants with a minor allele frequency (MAF) < 1% and three index variants are missense mutations (in genes ARSL, TSPAN6, and G6PD), all of which are novel. We validated the identified X chromosomal associations in up to 255,475 individuals from seven multi-ancestry cohorts (Additional file 21: Table S13). Twenty index variants were at least nominally associated (p-value < 0.05), with five reaching genome-wide signifi- cance in the replication cohorts alone (p-value < 5 × ­10−8, Additional file 32: Table S21). We additionally considered potential sex differences for the X chromosome variants. A missense variant in RENBP with MAF = 2.5% reached genome-wide significance only in males but was not significant in the sex-combined meta-analysis or in the female-only analysis (p-value = 4.59 × ­10−8, 0.003 and 0.2, respectively). We also observe three X chromosome loci with significant heterogeneity in effects between sexes; however, these were not significant in the replication cohorts alone, possibly due to the lower sample size (Bonferroni correction for the number of index variants in each trait, Additional file 32: Table S21). Using a PheWAS approach in the UK Biobank, we found four of the novel loci to have pleiotropic associations with body composition traits (FAM9B [HDL-C], EDA2R [HDL, TG], TSPAN6 [LDL-C, TC], and DOCK11 [HDL-C]), four variants with coronary ath- erosclerosis and ischemic heart disease, three with immune-related biomarkers (SLC9A7 [HDL-C], CLCN5 [HDL-C], THOC2 [HDL-C]), and two with blood clotting-related biomarkers (KLF8 [TG], TEX11 [HDL-C]) (Additional file 32: Table S21). Interestingly, two of the three sex-biased X chromosome variants demonstrate the most significant association with testosterone of all lipid X chromosome variants tested in the PheWAS (rs505520: beta/SE =  − 0.089/0.007 nmol/L per TG-increasing allele and rs5934507: beta/SE = 0.237/0.006 nmol/L per HDL-increasing allele). Discussion In this study, we identify and prioritize likely candidate genes at lipid-associated loci dis- covered through a variety of approaches including multi-ancestry meta-analysis of auto- somes [24] (~ 91 million variants) and the X chromosome (~ 3 million variants), as well as sex-specific meta-analyses using sample sizes ranging from 1.35 to 1.65 million indi- viduals. We previously reported a comparison of multi-ancestry vs single-ancestry lipid findings using autosomal chromosomes and identified improvements in fine-mapping of credible sets and PGS performance, with slight differences in the number of identi- fied loci by ancestry group [24]. Here, we add X chromosome and sex-specific discovery results. We also focus on lipid biology by prioritizing implicated genes, identifying novel phenotypes and diseases associated with genetically predicted lipid levels, and predict- ing candidate drug target genes. Our results from this effort translate our GWAS findings for three complimentary research areas, helping us further elucidate the biological mechanisms underlying the lipid-associated genetic variants. We first sought to identify methods for prioritization of functional genes at GWAS loci by performing six gene prioritization methods. Lipids are an excellent exemplar phenotype for gene prioritization algorithms because of a wealth of GWAS loci (~ 1000), Mendelian dyslipidemia genes (21), and mouse dyslipidemia phenotypes observed in gene knockouts (740). While the gene prioritization approaches are not independent of each other, integrating several prioritization predictors provides Kanoni et al. Genome Biology (2022) 23:268 Page 20 of 42 higher confidence when attempting to characterize causal genes. Others have also high- lighted the importance of such frameworks in different diseases [29, 54, 55]. We identify 466 unique genes by combining evidence from a global approach (PoPS) with local gene prioritization approaches. The vast majority of these genes had many lipid-related publications, suggesting the accuracy of our combined prioritization approach. Twenty-three PoPS + identified genes had no lipid-related publications, indi- cating they could be truly novel or possibly were incorrectly prioritized. Functional validation of the larger pool of prioritized genes, which will require highly parallel exper- imental methods, will help to further optimize bioinformatics algorithms to prioritize genes and is beyond the scope of this manuscript. Our prioritization approach also indicates several genes as potential drug targets including PDE3A and NR1H4. PDE3A encodes the phosphodiesterase 3A gene and is predicted to be druggable as phosphodiesterase 3A inhibitor (CILOSTAZOL). Cilosta- zol has antiplatelet, anti-proliferative, vasodilatory, and ischemic-reperfusion protective properties [56] and has been previously suggested for the primary or secondary preven- tion of CAD [22]. NR1H4 encodes a bile acid receptor and regulates the expression of genes involved in bile acid synthesis and transport. The target gene is predicted to be druggable as a bile acid receptor FXR agonist (URSODIOL). Ursodiol is used to treat primary biliary cirrhosis and cholelithiasis and could be a potential candidate for drug repurposing. We also identify eighteen coding variants where the protective lipid allele is also pro- tective for CAD or T2D. Among these, PCSK9 is a well-documented drug target, not only for lipids but also for cardiovascular events [57–59]. In comparison to published studies [60], others find a non-significant increased risk for T2D [61] and an arguably stronger protective effect for CAD [62], for PCSK9 variant carriers. Our observation is consistent with the lack of excess T2D risk observed in PCSK9 inhibitor clinical trials [57–59, 63] and with strong protective effects for coronary heart disease [64]. Further- more, these variants are potential therapeutic targets for protective lipid profiles and lowering risk of disease. Our second goal was to identify diseases that may benefit from lipid-lowering as well as diseases or traits that may become problematic due to very low lipids. To accom- plish this, we examined the association of genetically predicted lipid traits (using PGS) with 773 phenotypes in 478,556 individuals. We observed that genetically predicted increased LDL-C, TC, and HDL-C levels, or decreased TG levels, decrease the risk of cholelithiasis. Prior epidemiological studies have not consistently reported an associa- tion between lipid levels and risk of gallstones, with some studies showing that increased levels of LDL-C, TC, and TG and decreased levels of HDL-C predispose to the risk for cholelithiasis [65, 66], but others showing no association [67, 68]. Our results are cor- roborated by a recent Mendelian randomization meta-analysis study in the FinGen and UK Biobank cohorts [69]. The prioritized genes for the individual index lipid variants significantly associated with cholelithiasis in the PheWAS analysis include ABCG8, a hepatic cholesterol transporter, responsible for the efflux of cholesterol from the enterocytes to the lumen and from the hepatocytes into bile [70]. The lipid-decreasing allele of index variant in ABCG8, rs4245791, has been previously associated with high risk for gallstone disease [47] and high intestinal cholesterol absorption [71], possibly Kanoni et al. Genome Biology (2022) 23:268 Page 21 of 42 mediated by an increased expression of ABCG8 [72]. Furthermore, even after excluding ABCG5/8 variants from the LDL-PGS, the association with the risk of cholelithiasis was not attenuated. These PGS-PheWAS results suggest the existence of many other cho- lesterol transporters like ABCG8 that modify blood cholesterol levels perhaps in large part by facilitating an increased secretion of cholesterol into the biliary system, which in turn increases the risk of the formation of gallstones through the supersaturation of bile. We also observed that HbA1c levels were elevated among subjects with genetically predicted increased LDL-C and TC and with genetically predicted decreased HDL-C. Previous epidemiological studies have established associations between dyslipidemia (increased LDL-C, TC, TG, and decreased HDL-C levels) and increased HbA1c levels among subjects with T2D, as well as insulin-resistant subjects without diabetes [73, 74]. Our observations support a strong genetic basis to these associations and are in accord- ance with previous studies showing shared pathways between lipid biology, T2D, and HbA1c [75], as well as pleiotropic effects of blood red cell variants with lipid levels [76]. Mendelian randomization studies have shown that hemoglobin and LDL show bidirec- tional inverse relationships and hemoglobin effects on LDL are also mediated through Hb1Ac, implying that genetic variation influencing erythrocytic factors could also deter- mine lipid levels and the opposite [77]. While most of our significant PheWAS findings could be confirmed via Mendelian randomization studies, we cannot exclude the possi- bility of spurious associations due to pleiotropy. Lastly, we sought to expand the coverage of the genome and performed the most comprehensive GWAS of lipid levels to date by including assessment of 3 million variants on the X chromosome as well as explicitly testing for sex-specific effects across 23 chromosomes in 1.35 million individuals of diverse ancestries. We report 21 novel X chromosome loci, including an LDL-lowering locus involving a missense variant in G6PD, known to cause G6PD deficiency (p.V68M) [78]. The proposed mechanism is via the inhibition of the NADPH-dependent hydroxymethylglutaryl- CoA (HMG-CoA) reductase, resulting in decreased cholesterol biosynthesis, even though the protective effect of the G6PD deficiency on cardiovascular risk is debat- able [79]. We also observed that approximately 3–5% of the genome-wide lipid index variants exhibited differential effects between sexes, which did not seem to be due to differen- tial prevalence in the use of lipid medications or study selection bias. These findings may have important implications in the interpretation of lipid biology, the identification of novel drug targets, and possibly for more accurate prediction of blood cholesterol- related conditions. For example, the UGT2B17 locus, one of the ten sex-biased loci with corresponding sex-biased eQTL effect, is known to be implicated in androgen and drug metabolism [52]. A common CNV in the region, partially tagged by the lipid index vari- ant, is associated with significant variations in expression levels between ethnic groups [80], which would explain lack of replication in the set of independent studies, and the deletion has been linked to testosterone-related decreased BMI levels [81], as well as decreased risk for osteoporosis in men [82]. Several of the reported sex-biased and X chromosome loci showed significant pleiotropic effects with sex hormone levels, including testosterone and SHBG, high- lighting the role of hormone regulation in lipid metabolism [83]. In particular, we Kanoni et al. Genome Biology (2022) 23:268 Page 22 of 42 observe an overall inverse effect between the X chromosome lipid index variants and the sex hormone levels. Observational studies have long suggested a potential influ- ence of the sex hormones on the risk for cardiovascular risk [84] but this hypothesis has not been consistently supported by recent Mendelian randomization studies, raising the issues of reverse causality [85, 86]. Conclusions In conclusion, we leverage the power of a large multi-ancestry GWAS study to fur- ther our understanding of lipid metabolism and the impact on chronic diseases. We identify novel lipid loci on the X chromosome and autosomal loci with evident sex- biased lipid effects. We compare a range of gene prioritizing methods to identify causal genes, an approach applicable to studying other complex traits. We addition- ally further our understanding of lipid metabolism through a phenome-wide study that implicates a relationship between genetically predicted low cholesterol with risk of cholelithiasis. Methods Meta‑analysis Summary statistics for sex-combined autosomal analyses were previously pub- lished [24]. Following the same procedure, we carried out meta-analyses stratified by sex for 5 lipid traits (HDL-C, LDL-C, TG, nonHDL-C, and TC) for both the auto- somes and chromosome X. The sample size for chromosome X (Total N = 1,238,180; males = 749,391; females = 562,410) was lower than available for autosomes as not all participating biobanks submitted results for chromosome X. Quality control of sum- mary statistics from contributing cohorts was performed using EasyQC [87]. Prior to meta-analysis, we removed variants with low imputation info scores (r2 < 0.3), those with minor allele count < 3, and those with Hardy–Weinberg equilibrium p-value < 1 × ­10−8. Variants on the X chromosome were filtered using the female imputation info scores and Hardy–Weinberg equilibrium p-values. Summary statis- tics were corrected by the genomic-control factor calculated from the median p-value of variants with minor allele frequency > 0.5%. For cohorts that contributed summary statistics imputed both on the Haplotype Reference Consortium (HRC) and 1000 Genomes Population v3 (1KGP3) panels, we generated a single file containing all pos- sible variants, favoring those imputed from the HRC imputation panel due to gen- erally higher imputation quality of these variants. Multi-ancestry meta-analysis was performed with MR-MEGA [88] with 5 principal components and using the inverse- variance weighted method in METAL to estimate effect sizes [89]. Independent loci were defined with physical distance > 500 kb or genetic distance > 0.25 cM, whichever one would result in a larger window, followed by a conditional analysis using rareG- WAMA [90] as previously described [24], to identify index variants that were shad- ows of nearby, more-significant associations. Conditional analysis for chromosome X used a female-only UK Biobank LD reference (N = 21,510). In line with the analysis in the autosomes, a locus was identified as dependent if the effect size after conditioning Kanoni et al. Genome Biology (2022) 23:268 Page 23 of 42 on the most significant variant in the area was more than 1.43 times smaller than the original (95th percentile of the effect size ratios for chromosome X). Differences in effect size between males and females were tested within each cohort using [91]: Bm − Bf Z= se2m + se2f − 2 ∗ r ∗ sef ∗ sem and were then meta-analyzed across studies using METAL, to account for cohort-spe- cific ascertainment (e.g., enrichment of cases for type 2 diabetes), or demographics, such as age. Replication We collected summary statistics from 8 cohorts across 6 ancestry groups, includ- ing African or African American, East Asian, European, Hispanic, Middle Eastern, and South Asian. Each cohort provided sex-stratified and X chromosome association results for the tested traits, as available. The difference in effect sizes between males and females was calculated within each cohort as described above and then meta- analyzed across studies using METAL. X chromosome association results were meta- analyzed using METAL with weighting by sample size. Gene prioritization methods Closest gene We annotated the closest gene to the lipid multi-ancestry index variants [24] by identify- ing the closest gene transcript on either side (500 kb) of the index variant [92]. Colocalization with GTEx eQTLs For each of the five lipid phenotypes, we first lifted over GWAS summary statistics from the multi-ancestry meta-analysis [24] to GRCh38 using the UCSC liftOver tool. Then, we defined a set of approximately independent windows across the genome within which colocalization with eQTLs was run. To define these, we first identified all genome- wide significant variants (p-value < 5e − 08) from the meta-analysis for each lipid trait and sorted them by significance, from most significant to least. Starting with the most significant variant, we aimed to define a window defining independent genetic signals; we define a variant’s window as a region within the greater of 500 kb or 0.25 cM on either side of this “sentinel variant.” Genetic distances were defined using reference maps from HapMap 3. All other genome-wide significant variants within this window were discarded from the list of sentinel variants, and similar windows were defined for the remaining genome-wide significant variants. We ran an eQTL colocalization using GTEx v8 eQTL summary statistics within each of our defined windows for all lipid traits. For each of the 49 GTEx tissues, we first iden- tified all genes within 1 Mb of the sentinel variant, and then restricted analysis to those genes with eQTLs (“eGenes”) in that tissue (FDR < 0.05). We used the R package “coloc” Kanoni et al. Genome Biology (2022) 23:268 Page 24 of 42 (run on R version 3.4.3, coloc version 3.2.1) [93] with default parameters to run colo- calization between the GWAS signal and the eQTL signal for each of these cis-eGenes, using as input those variants in the defined window, i.e., all variants present in both datasets. A colocalization posterior probability of (PP3 + PP4) > 0.8 was used to identify loci with enough colocalization power, and PP4/PP3 > 0.9 was used to define those loci that show significant colocalization [94]. Transcriptome‑wide association studies (TWAS) For our transcriptome-wide association analysis (TWAS), we integrated the results of our GWAS with eQTL summary statistics from GTEx v8. The S-PrediXcan software [95] allows us to integrate these two datasets using only summary statistics from GWAS without needing individual-level genotype data. We used the multi-ancestry lipid GWAS summary statistics [24] and harmonized them with the GTEx summary statistics. Then we performed the TWAS using the eQTL models estimated on GTEx v8 expression data. For each of the 49 GTEx tissues, we identified “significant genes” those genes with p-values more significant than an FDR threshold of 0.05. Genes with coding variants We determine the coding variants within 99% credible sets and the coding variants in LD > 0.8 with variants in the 99% credible sets with the credible sets as defined here [24]. We define regions for construction of the credible sets as ± 500 kb around each index variant. We used Bayes factors (BFs) for each variant from the MR-MEGA output and generated the credible sets within each region by ranking all variants by BF and cal- culating the number of variants required to reach a cumulative probability of at least 99%. Additionally, we used previously established gene-based associations [96] to deter- mine whether rare coding variation in a gene were associated with blood lipid levels (p < 0.001). We labeled a gene as having coding variants if any of these criteria were met. DEPICT We used Data-driven Expression-Prioritized Integration for Complex Traits (DEPICT, v1 beta version rel194 for 1 KG imputed GWAS) to prioritize genes at our index vari- ants, on the assumption that truly associated genes share functional annotations [27]. Index variants [24] with p-value < 5 x ­10−8 were retained as input. We implemented the DEPICT analysis with the default settings of 500 permutations for bias adjustment and 20 replications for FDR estimation. DEPICT prioritizes genes by calculating the similar- ity of a given gene to genes from other associated loci across 14,461 reconstituted gene sets and estimates the nominal gene prioritization p-value and the estimated false dis- covery rate of each gene. The prioritized genes at FDR < 0.05 were considered significant. PoPS We used the PoPS method to prioritize genes in the previously reported [24] multi- ancestry index variants for all lipid traits. The PoPS method [28] is a new gene prioriti- zation method that identifies the causal genes by integrating GWAS summary statistics with gene expression, biological pathway, and predicted protein–protein interaction Kanoni et al. Genome Biology (2022) 23:268 Page 25 of 42 data. First, as part of the PoPS analysis, we used MAGMA to compute gene associa- tion statistics (z-scores) and gene–gene correlations from GWAS summary statistics and LD information from a multi-ancestry reference panel (1000 Genomes). Next, PoPS performs marginal feature selection by using MAGMA to perform enrichment analy- sis for each gene feature separately. The model is fitted by generalized least squares (GLS), and MAGMA results are used to perform marginal feature selection, retaining only features that pass a nominal significance threshold (p < 0.05). Then PoPS computes a joint enrichment of all selected features simultaneously in a leave one chromosome out (LOCO) framework. The gene features employed by PoPS are listed here: https://​github.​ com/​Finuc​aneLab/​gene_​featu​res. Finally, PoPS computes polygenic priority scores for each gene by fitting a joint model for the enrichment of all selected features. The PoPS score for a gene is independent of the GWAS data on the chromosome where the gene is located. The PoPS analysis returned scores for a total of 18,383 genes per lipid trait. We only kept the top 20% genes among all 18,383 genes. We then annotated our index vari- ants with the nearest ENSEMBL genes in a 500-kb window (either side) and selected the highest PoPS score gene in the locus as the prioritized one. We performed the PoPS analysis on our lipid-specific multi-ancestry meta-analysis results, using all populations from 1000G as the reference for the LD information in MAGMA. As a sensitivity step, we also repeated the same analysis using only the Euro- pean population from 1000G as the reference. We observed high concordance in the top two PoPS prioritized genes from both reference panels. In detail, the same 2119 genes (89%) were prioritized as the top genes from both panels, a further 203 genes were prior- itized as a top gene with one panel and as the second top with the other and only 7 genes were completely mismatched between the two reference panels. Monogenic genes We annotated genes known to cause Mendelian lipid disorders based on proximity with identified GWAS loci [97, 98]. GWAS index variants within ± 500 kb of the transcription start and end positions from the USCS genome browser annotations were annotated as nearby known monogenic dyslipidemia genes. Mouse knockout lipid phenotype silver set genes Human gene symbols (9557 unique genes) were mapped to gene identifiers (HGNC) and their corresponding mouse ortholog genes were obtained using Ensembl (www.​ ensem​bl.​org). Phenotype data for single-gene knockout mouse models were obtained from the International Mouse Phenotyping Consortium (IMPC) (www.​mouse​pheno​ type.​org) latest data release 12.0 (www.​mouse​pheno​type.​org/​data/​relea​se). The knock- out mouse models were primarily produced by IMPC but also include some models that have been reported from the relevant literature and were curated by Mouse Genome Informatics (MGI) data release 6.16 (www.​infor​matics.​jax.​org). For each mouse model, reported phenotypes were grouped using the mammalian phenotype ontology hierar- chy into broad categories relevant to lipids: growth and body weight (MP:0001259), lipid homeostasis (MP:0002118), cholesterol homeostasis (MP:0005278), and lipid metabo- lism (MP:0013245). This resulted in mapping of human genes to significant phenotypes in animals. Kanoni et al. Genome Biology (2022) 23:268 Page 26 of 42 For each of the multi-ancestry lipid index variant [24], we mapped the closest gene to the knockout mouse phenotypes and curated the set to only include mouse pheno- types strictly relating to lipid metabolism. That resulted in our silver set of 740 genes with mouse lipid phenotypes (Additional file 33: Table S22). Overlap between methods We standardized the gene names across different methods using the R/geneSynonym package, a wrapper to gene synonym information in ftp://​ftp.​ncbi.​nlm.​nih.​gov/​gene/​ DATA/​gene_​info.​gz. We also quantified how often the same gene was prioritized by multiple methods for each index variant and determined scores that ranged from 1 to 6 (S1-S6), based on the number of methods that prioritized the gene. We integrated multiple gene prioritization methods to identify likely causal genes in the latest global lipid genetics consortium GWAS results. In total, we have implemented the 6 individual gene prioritization methods above that utilize the GWAS summary statistics from meta-analysis. Our gene prioritization methods can be placed into two broad categories, the locus-specific methods and the genome-wide methods. The locus- specific methods leverage local GWAS data by connecting the causal variants to the causal gene(s) using genomic distance, eQTLs, or protein-coding variants. More specifically, there are four locus-specific methods that have been implemented including: (1) The closest protein-coding gene around the index variants based on the genomic distance, (2) eQTL colocalization using r COLOC package, (3) TWAS using S-PrediXcan, (4) coding variants which have been identified in 99% credible sets OR in LD > 0.8 with coding variants OR from gene-based tests (p < 0.001) of rare coding vari- ants. For the eQTL and TWAS, we first used all the 49 GTEx tissues and then restricted to only 5 lipid-specific tissues: liver, adipose subcutaneous, adipose visceral, whole blood, and small intestine. In addition, two genome-wide methods were employed: (1) DEPICT (FDR < 0.05), (2) PoPS (Top 1 gene). It is reasonable to combine similarity-based methods with locus-based methods since they use two different sources of information. To determine the relative performance of each prioritization method and their com- bined scores for lipid loci, we used 21 genes known to cause Mendelian dyslipidemias as a gold standard set (ABCA1, ABCG5, ANGPTL3, APOA5, APOB, APOE, CETP, CYP27A1, GPD1, GPIHBP1, LCAT​, LDLR, LDLRAP1, LIPA, LIPC, LMF1, LPL, MTTP, PCSK9, SAR1B, SCARB1), and 740 mouse knockout genes causing lipid phenotypes as a silver standard set (Additional file 33: Table S22). We examined two metrics for each gene prioritization approach: (1) the proportion of prioritized genes in the gold/silver standard set, and (2) the proportion of correctly identified genes among all prioritized genes (Additional file 3: Figure S1). Note that out of the 2286 lipid associations, 97 fell within 500 kb of a Mendelian gene and 1280 within 500 kb of a mouse knockout gene with a lipid phenotype. We observed that the TWAS results yielded a high number of prioritized genes, but lead to a low proportion correctly identified. The TWAS approach had a much smaller proportion of genes correctly prioritized among all the prioritized genes, given it prioritized a total of 3511 genes, which was 3.5-fold greater than the other methods (~ 1000 genes). Notably, PoPS provided a similar proportion of correctly identi- fied genes (78%) as of TWAS, while retained a high proportion of prioritized genes in the gold standard set (67%). Compared with PoPS, PoPS + (PoPS plus one of the local Kanoni et al. Genome Biology (2022) 23:268 Page 27 of 42 methods) slightly sacrificed the proportion of correctly identified genes from 78 to 71%, but improved the proportion of prioritized genes in the gold standard set from 67% to 79%. Overall, PoPS/PoPS + outperform other gene prioritization methods on both metrics for our gold (Additional file 3: Figure S1A) and silver (Additional file 3: Figure S1B) standard gene sets. We also assessed lipid-relevant tissue (liver, subcutaneous and visceral adipose, whole blood, and small intestine) expression QTLs (lipid eQTLs) and transcriptome-wide association (lipid TWAS) and found that the expression results from all tissues performed slightly better at recovering the reference gene sets compared with limiting to the lipid-relevant tissues (Additional file 3: Figure S1). Text mining analysis We retrieved the whole MEDLINE/PubMed titles and abstracts as of March 06, 2022, from National Library of Medicine (https://​ftp.​ncbi.​nlm.​nih.​gov/​pubmed/​basel​ine/; https://​ftp.​ncbi.​nlm.​nih.​gov/​pubmed/​updat​efiles/). We then examined whether a list of genes prioritized by PoPS + and any one of the lipid-related keywords (lipid, lipids, triglyceride, triglycerides, fatty acid, cholesterol, dyslipidemias, hyperlipidemia, hyper- cholesteremia, diabetes, type 2 diabetes, type II diabetes, heart, cardiovascular, artery, coronary, coronary artery, coronary heart, atherosclerosis, peripheral vascular, PAD, stroke) occurred in the same abstract. We counted how many lipid-related publications that have a specific gene co-occurred with at least one lipid-related keyword. The same text mining approach was also implemented to a set of randomly selected genes from the 18,383 protein-coding genes used by the PoPS. We estimated the number of lipid-related publications we would expect to see by chance. A Mann–Whitney U test was performed to show whether there was a significant difference between the number of lipid-related publications of the PoPS + gene set and reference gene set. Drug target mining analysis To gain therapeutic insights from our gene prioritization results, we performed a lookup in Therapeutic Target Database (TTD) 2022 [99] (http://​db.​idrbl​ab.​net/​ttd/). Specifically, we cross-referenced 466 unique lipid-associated genes prioritized by PoPS + (Additional file 2: Table S2) with 1563 genes corresponding to at least one drug (either under devel- opment or approved) with known clinical indication in TTD 2022. As a quality control for this lookup, we excluded all TTD entries related to drugs that were discontinued, terminated, or withdrawn from the market. The full lookup results are available in Addi- tional file 8: Table S6. Driver tissues for lipid levels We performed phenotype-tissue association analysis using DESE (driver-tissue estima- tion by selective expression) [40]. DESE estimates the causal tissues by selective expres- sion of phenotype-associated genes in GWAS. We used the GWAS summary statistics from the five lipid traits and the GTEx v8 normalized gene-level and transcript-level expression datasets as input. SNPs inside a gene and its ± 5 kb adjacent regions were first mapped to the gene, and then DESE ran iteratively to produce a list of driver tissues and the corresponding p-values of the associations. We used a Bonferroni-corrected signifi- cance threshold of 0.05/54 = 9.3 x ­10−4. Kanoni et al. Genome Biology (2022) 23:268 Page 28 of 42 PheWAS analysis Construction of lipid PGSs We had previously developed a multi-ancestry PGS for LDL-C that was demonstrated to perform well across multiple ancestry groups [24]. In a similar manner, we also gen- erated PGS for HDL-C, nonHDL-C, TC, and triglycerides. First, multi-ancestry meta- analysis results were generated with METAL [89] after excluding individuals from the Michigan Genomics Initiative and the UK Biobank. The set of variants used to construct the PGS was limited to those that were well-imputed (R2 > 0.3) in MGI, UK Biobank, and MVP. Risk scores based on PRS-CS [100] or pruning and thresholding with Plink [101] across several r2 (0.1, 0.2), distance (250 kb, 500 kb), and p-value thresholds (5 × ­10−10, 5 × ­10−9, 5 × ­10−8, 5 × ­10−7, 5 × ­10−6, 5 × ­10−5, 5 × ­10−4, 5 × ­10−3, 0.05) were developed. For each trait, the single best score was selected based on the adjusted r2 calculated in the UK Biobank of the linear model for the lipid trait with the risk score and age, sex, batch, and PC1-4 as covariates. This corresponded to PRS-CS for HDL-C and non- HDL-C and pruning and thresholding for LDL-C (r2 = 0.1, p-value = 5 × ­10−4, 500 kb), TG (r2 = 0.1, p-value = 5 × ­10−3, 500 kb), and TC (r2 = 0.1, p-value = 5 × ­10−4, 500 kb). The variance explained by the risk score among the UK Biobank participants was simi- lar across traits (adjusted r2 of the full model-adjusted r2 of covariates: HDL-C = 0.13; LDL-C = 0.15; nonHDL-C = 0.14; TC = 0.14; TG = 0.10) and validated the ability of the risk score to predict genetically increased lipid levels. PheWAS of lipid PGSs and index lipid variants in the UK Biobank and MVP We used the European ancestry subset of individuals from the UK Biobank (408,886 samples) and the European samples from MVP (69,670 samples) to perform the PheWAS analysis. We constructed a weighted PGS for each of the lipid traits, based on the correspond- ing genome-wide significant multi-ancestry index variants. We used the PheWAS package in R [102] to map ICD-10 codes from hospital records into clinically relevant phenotypes (phecodes) and to implement these association analyses, while adjusting for sex, age, 10 genetic principal components, and genotyping array (for the UK Biobank only) in each cohort. For the lipid-PGS PheWAS, each PGS was inverse normalized prior to analysis and lipid levels were corrected for statin use. The MVP samples used for the PheWAS analysis were not included in the GWAS meta-analysis [24]. Similarly, we extracted all multi-ancestry autosomal index variants for all lipid traits from the same European ancestry subset of the UK Biobank and MVP and performed a single-variant PheWAS association analysis per cohort. Additionally, we performed a single-variant PheWAS association analysis in the UK Biobank only with the sex-strati- fied and X chromosome index variants from the multi-ancestry analysis. Meta‑analysis of MVP and the UK Biobank PheWAS results We combined, via meta-analysis, PheWAS lipid-specific PGS results for all intersect- ing phecodes and biomarkers between the UK Biobank and MVP (Europeans only) per lipid trait. We used ICD10-based phecodes and manually matched biomarkers to iden- tify intersecting phenotypes between the two datasets. We restricted our meta-analy- sis to phenotypes that had at least 100 samples (total number for continuous traits or Kanoni et al. Genome Biology (2022) 23:268 Page 29 of 42 number of cases for binary traits) in each cohort. After the meta-analysis, we excluded phenotypes that had less than 500 combined samples (total number for continuous traits or number of cases for binary traits), to avoid reporting spurious results [103]. That resulted in a total of 773 phenotypes (739 phecodes and 34 biomarkers/measurements). We used both fixed and random effects model for the meta-analysis. We assessed het- erogeneity using the p-value for Cochran’s q and set the level for significant heteroge- neity at a Bonferroni threshold (p-value ≤ 6.5 × ­10−5, to account for multiple testing of 773 phenotypes). We report the results from the fixed-effects model for the phenotypes with non-significant heterogeneity and the results from the random effects model for all others. Similarly, we meta-analyzed all index-variant PheWAS results between the UK Biobank and MVP and obtained results for 811 phenotypes and 1750 lipid multi-ances- try index variants, after excluding instances with a combined sample size < 500. Lipid index variants with CAD, T2D, and NAFLD datasets The GWAS meta-analysis results of CAD and T2D were acquired from MVP [62] and DIAGRAM Consortium [61], respectively. For variant rs1229984, the CAD result is from CARDIoGRAMPlusC4D meta-analysis [104], as it was not present in the MVP results. The NAFLD GWAS and meta-analysis was performed in the UK Biobank and Michigan Genomics Initiative (MGI). We determined the association of the lipid index variants with CAD, T2D, and NAFLD and aligned the alleles across all the traits to the LDL-lowering allele. We then highlighted the protective lipid coding alleles associated with CAD. GWAS and meta‑analysis of NAFLD in the UK Biobank and Michigan Genomics Initiative (MGI) Individuals with NAFLD were identified using ICD-9 571.8 and ICD-10 K76.0. Indi- viduals with hepatitis, liver cirrhosis, liver abscess, ascites, a liver transplant, hepato- megaly, jaundice, or with abnormal result of serum enzyme levels or a function study of the liver were excluded (exclusion phecodes 70.2, 70.3, 571.51, 571.6, 571.8, 571.81, 572, 573, 573.2, 573.3, 573.5, 573.7, 573.9) [105]. Analysis was performed using SAIGE v43.3 [106]. Analysis in the UK Biobank included white British individuals with batch, sex, birth year, and the first 4 genetic principal components as covariates. A total of 1122 cases and 399,900 controls were included in the analysis. Analysis in MGI included only European-ancestry participants with array version, sex, birth year, and the first 4 genetic principal components as covariates. A total of 2901 cases and 49,098 controls were ana- lyzed. Meta-analysis was performed using METAL with weighting based on the effective sample size calculated as 4/((1/Ncases) + (1/Ncontrols)). CAD/T2D colocalization analysis with lipid traits We used R package coloc v3.2.1 [93] to perform summary statistics-based colocalization via a Bayesian approach and test whether the 5 lipid traits share common genetic causal variants with CAD or T2D. We first defined a window of ± 100 kb around each index variant [24]. Then for each window of the 10 pairs of traits, we ran colocalization with default parameters using those SNPs present in both datasets. A colocalization posterior probability of PP4 > 0.8 was used to define those loci that show significant colocalization. Kanoni et al. Genome Biology (2022) 23:268 Page 30 of 42 Supplementary Information The online version contains supplementary material available at https://​doi.​org/​10.​1186/​s13059-​022-​02837-1. Additional file 1: Table S1. Characteristics of contributing cohorts (as provided by each participating cohort). Additional file 2: Table S2. Association results for the multi-ancestry index SNPs with the gene prioritization. Additional file 3: Figure S1. Summary of prioritizing genes for A. Mendelian and B. mouse model genes separately by trait. Additional file 4: Table S3. Text mining results for the PoPS+ prioritized genes. Additional file 5: Table S4. Frequency of lipid-related publications for the PoPS+ prioritized genes. Additional file 6: Figure S2. Frequency distribution of the lipid-related publications for both high confidence genes and the baseline genes. Additional file 7: Table S5. Lookup of all prioritized lipid genes in the Therapeutic Target Database 2022. Additional file 8: Table S6. Association of lipid index variants with CAD, T2D and NAFLD. Additional file 9: Figure S3. Lipid traits – tissue/cell type associations estimated by DESE according to GTEx gene- level and GTEx transcript-level selective expression. Additional file 10: Table S7. DESE phenotype-tissue association results using both GTEx gene-level and transcript- level selective expression. Additional file 11: Figure S4. Comparison of PheWAS results in UKB and MVP for the LDL-C PGS, HDL-C PGS, TC PGS, TG PGS and nonHDL-C PGS. Additional file 12: Table S8. PheWAS UKB-MVP meta-analysis results for each lipid PGS. Additional file 13: Figure S5. PheWAS meta-analysis results for the trans-ethnic HDL-C PGS in UK Biobank and MVP. Additional file 14: Figure S6. PheWAS meta-analysis results for the trans-ethnic TC PGS in UK Biobank and MVP. Additional file 15: Figure S7. PheWAS meta-analysis results for the trans-ethnic TG PGS in UK Biobank and MVP. Additional file 16: Figure S8. PheWAS meta-analysis results for the trans-ethnic nonHDL-C PGS in UK Biobank and MVP. Additional file 17: Table S9. PheWAS UKB-MVP meta-analysis results for each index lipid variant at Bonferroni threshold for multiple testing p<=3.5e-8) Additional file 18: Table S10. Lambda GC values across minor allele frequency bins for sex-specific meta-analyses. Additional file 19: Table S11. Significant female-specific multi-ancestry meta-analysis results. Additional file 20: Table S12. Significant male-specific multi-ancestry meta-analysis results. Additional file 21: Table S13. Characteristics of replication cohorts (as provided by each participating cohort). Additional file 22: Table S14. Test for difference in effects for index variants from sex-stratified meta-analysis. Additional file 23: Table S15. Comparison of the sex-specific effects. Additional file 24: Figure S9. Comparison of effect size estimates between males and females for index variants showing a significant difference in effect size between sexes. Additional file 25: Table S16. Comparison of effect size estimates for sex-stratified analysis in the replication cohorts. Additional file 26: Figure S10. Comparison of effect sizes for trans-ancestry index variants excluding cholesterol- lowering medication. Additional file 27: Table S17. Sex-stratified effect sizes in UK Biobank considering all individuals or only those not on cholesterol lowering medications. Additional file 28: Table S18. Sex-participation association of the variants with significant sex-specific lipid results. Additional file 29: Table S19. Comparison of sex-stratified effect sizes in the UK Biobank for BMI, waist hip ratio adjusted for BMI, alcohol use, and smoking status. Additional file 30: Table S20. Colocalization results for the sex-specific loci. Additional file 31: Supplementary Note. Supplementary Note and Cohort Acknowledgments. Additional file 32: Table S21. Significant X chromosome results in the sex-combined and sex-stratified analysis and replication. Additional file 33: Table S22. Mouse genes with lipid phenotypes (silver set). Additional file 34. Review history. Acknowledgements We thank Bethany Klunder for her administrative support of the Global Lipids Genetics Consortium. Study-specific acknowledgements are available in the Additional file 32: Supplementary Note. Kanoni et al. Genome Biology (2022) 23:268 Page 31 of 42 Peer review information Wenjing She was the primary editor of this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Review history The review history is available as Additional file 34. Authors’ contributions S.Kanoni, S.E.G. Y.W., I.S., S.Ramdas, and Xiang.Zhu contributed equally to this work as co-first authors. All authors reviewed the manuscript. Consortium management: G.M.P., P.N., T.L.A., M.Boehnke, and C.J.W. Study design, interpreta‑ tion of results, and drafting of the manuscript: S.Kanoni, S.E.G., Y.W., I.S., S.Ramdas, Xiang.Zhu, S.L.C., K.F.B., S.Vedantam, T.W.W., A.E.L., E.M., G.JM.Z., K-H.H.W., I.N., Y.V.S., A.P.M., M.Boehnke, C.D.B., P.N., P.D., C.J.W., T.L.A., and G.M.P. Primary analyses: S.Kanoni, S.E.G., Y.W., I.S., S.Ramdas, Xiang.Zhu, S.L.C., K.F.B., S.Vedantam, T.W.W., A.E.L. Individual study design, analysis, and oversight: S.Kanoni, S.E.G., Y.W., I.S., S.Ramdas, Xiang.Zhu, S.L.C., K.F.B., S.Vedantam, T.W.W., A.E.L., E.M., G.JM.Z., K-H.H.W., I.N., Q.H., D.K., A.T.Hilliard, Zeyuan.Wang, C.X., G.Thorleifsson, A.H., D.F.G., H.Holm, I.O., M.Y.H., S.Han, M.Akiyama, S.S., C.Terao, M.Kanai, W.Zhou, B.M.B., H.R., A.S.H., Y.V., J.A.P., E.A.R., T.Lingren, QP.F., I.J.K., A.N., J.Takayama, H.C.M., K.A.H., B.T., J.Haessler, F.G., Y.B., J.E.M., A.Campbell, K.L., I.Y.M., A.R., G.Hindy, J.D.F., W.Zhao, D.R.W., C.Turman, H.Huang, M.Graff, A.Choudhury, D.Sengupta, A.Mahajan, M.R.B., W.Zhang, K.Yu, E.M.S., A.Pandit, S.G., X.Y., J.Luan, J-H.Z., F.Matsuda, H-M.J.,.Yoon, C.M-G., A.Pitsillides, J.J.H., A.R.Wood, Y.J., Z.G., S.Haworth, R.E.M., J.F.C., M.Aadahl, A.A.B., J.Yao, A.Manichaikul, C-M.H., Y-J.H., H.R.W., J.R., J.B-J., L.L.K., A.G., M.S-L., R.N., P.Mauro, F.Matteo, A.F.McD., P.M-V., M.Wielscher, S.T., N.S., L.T.M., M.Munz, L.Z., J.Huang, B.Y., A.Poveda, A.K., C.Lamina, L.F., M.S., T.E..G., J.P.B., S.E.R., E.W.D., J.M.Z., J.S.M., C.Fuchsberger, H.Christensen, J.A.B., M.V-M., M.F.F., M.K.W., Zhe.Wang, M.H.P., M.Mangino, P.C., N.V., J.W.B., J.Engmann, N.L.T., A.V., R.C.S., K.S.L., N.R.Z., P.L., M.E.K., G.E.D., S.Huo, D.D.I., H.I., Jian.Yang, Jun.Liu, A.D., H.L.L., J.M., M.Frank, B.S., L.J.S., M.CG., C.W., M.Nakatochi, A.W., N.H–K., X.S., R.X., A.H-C., J.C.F-L., V.L., S.S.N., S.Bayyana, H.M.S., M.R.I., C.Oldmeadow, H-N.K., S.Ryu, P.RHJ.T., L.A., R.D., L.A.L., G.Prasad, L.L-M., M.P., J.Long, X.Li, E.T., F.T., C.N.S., A.Loukola, S.Bollepalli, S.C.W., Y.X.W., W.B.W., T.Nutile, D.R., Y.J.S., S.C., F.L., Jingyun.Yang, K.A.K., B.B., G.G.N., K.M., L.F.B., J.A.S., P.H., A-E.F., E.H., M.Lin, M.P.C., S.Vaccargiu, P.J.van der M., N.Pitkänen, B.E.C., S.W.van der L., K.N.C., S.W., A.R.B., A.P.D., A.A.A., J.Y.L., E.RB.P.,.A.N., H.S.C., M.Nethander, S.F-W., L.S., N.W.R., C.A.W., S-Y.L., J-S.W., C.C., L-P.L., K.N., G.C-P., H.Vestergaard, B.H., O.G., Q.C., M.O.O., J.van S., J.Liang, H.T., N.T., J.H.S., R.D.J., A.P.R., L.W.M., Z.C., L.Li, T.Kawaguchi, J.Thiery, J.C.B., L.J.L., Huaixing.Li, M.A.N., O.T.R., S.I., S.H.W., C.P.N., H.Campbell, S.J., T.Nabika, F.A-M., H.N., P.S.B., I.K., P.K., T.G., T.Katsuya, D.de K., Gert J.de B., E.K.K., H.H.H.A., M.A.I., Xiaofeng.Zhu, F.W.A., A.O.K., J.WJ.B., X-O.S., L.S.R., O.Pedersen, T.H., P.Mitchell, A.W.H., M.Kähönen, L.P., C.Bouchard, A.T., Y-D. I.C., C.E.P., T.A.M., W.L., A.F., C.Ohlsson, D.M., Y.S.C., H.Lee, J-M.Y., W–P.K., S.Y.R., J-T.W., I.M.H., K.J.S., M.E.Z., H.Völzke, G.Homuth, M.K.E., A.B.Z., O.Polasek, G.Pasterkamp, I.E.H., S.Redline, K.P., A.J.O., H.Snieder, G.B., R.S., H.Schmidt, S.Bandinelli, G.D., T.A.T., S.LR.K., P.A.P., N.K., M.B.S., G.G., C.A.B., B.J., P.K.J., D.A.B., P.L.De J., X.Lu, V.M., M.Brown, M.J.C., P.B.M., X.G., M.Ciullo, J.B.J., N.J.S., J.Kaprio, P.P., T.T-L., C.A.A-S., L.S.A., S.A.B., H. J.de S., A.R.Wickremasinghe, R.M.K., J-Y.W., W.Zeng, A.I.den H., D.B., A.Correa, J.G.W., L.Lind, C-K.H., A.E.N., Y.M.G., J.F.W., B.P., H–L.K., J.A., R.J.S., D.C.R., D.K.A., M.Walker, H.A.K., G.R.C., J.M.M., M.C.C., D.J., N.P.B., C.G.V., L.O., M.Fornage, E S.T., R.M.van D., T.Lehtimäki, N.C., M.Yokota, Jianjun.Liu, D.F.R., A.J.McK., F.Kee, K-H.J., M.I.McC., C.NA.P., V.V., C.H., E.S., C.M.van D., Z-B.J., J.Q., H.Hishigaki, X.Lin, W.M., V.G., J-C.T., G.L., L.M.t H., P.JM.E., S.M.D., M.Kumari, M.Kivimaki, P.van der H., T.D.S., R.J.F.L, M.A.P., E.J.P., M.Cruz, B.M.P., I.B., P.P.P., C.N.R., K.Christensen, S.Ripatti, E.W., H.Hakonarson, S.F.A.G., L.ALM.K., J.de G., M.Loeffler, F.Kronenberg, D.G., J.Erdmann, H.Schunkert, P.W.F., A.Linneberg, J.W.J., A.V.K., M.Männikkö, M-R.J., Z.K., C.Francesco, D.O.M–K., K.W.van D., H.W., D.P.S., N.G., P.S., N.Poulter, L-M.C., J.I.R., T.M.D., F.Karpe, M.J.N., N.J.T., C-Y.C., T-Y.W., C.C.K., Hengtong.Li, C.S., A.Peters, C.G., A.T.Hattersley, N.L.P., P.KE.M., D.I.B., A.HM.W., L.A.C., J.B.J.vanM, M.Ghanbari, P.G-L., W.H., YJ.K., Y.T., N.J.W.,.Langenberg, E.Z., J.Kuusisto, M.Laakso, E.I., G.A., J.C.C., J.S.K., P.S.de V., A.C.M., S.Hazelhurst, M.R., K.E.N., M.D., P.K., N.G.M., J.B.W., S.A., D.Saleheen, R.G.W., M.V.H., C.Black, B.H.S., A.B., A.E.J., J.E.B., P.M.R., D.I.C., C.K., G.Tamiya, M.Yamamoto, D.A.van H., R.C.T., W-Q.W., G.P.J., B.N., M.G.H., M.D.R., P.J., V.S., K.H., B.O.Å., M.Kubo, Y.K., Y.O., Y.M., B-J.K., U.T., K.S., J.Z., Y.E.C., Y-L.H., J.A.L., D.J.R., P.S.T., K-M.C., K.Cho, C.J.O’D., J.M.G., P.WF.W., T.M.F., J.N.H., S.Kathiresan, K.L.M., Y.V.S., A.P.M., M.Boehnke, C.D.B., P.N., P.D., C.J.W., T.L.A., and G.M.P. All author(s) read and approved the final manuscript. Funding GMP, PN, and CW are supported by NHLBI R01HL127564. GMP and PN are supported by R01HL142711. AG acknowledge support from the Wellcome Trust (201543/B/16/Z), European Union Seventh Framework Programme FP7/2007–2013 under grant agreement no. HEALTH-F2-2013–601456 (CVGenes@Target) & the TriPartite Immunometabolism Consortium [TrIC]-Novo Nordisk Foundation’s Grant number NNF15CC0018486. JMM is supported by American Diabetes Associa‑ tion Innovative and Clinical Translational Award 1–19-ICTS-068. SR was supported by the Academy of Finland Center of Excellence in Complex Disease Genetics (Grant No 312062), the Finnish Foundation for Cardiovascular Research, the Sigrid Juselius Foundation, and University of Helsinki HiLIFE Fellow and Grand Challenge grants. EW was supported by the Finnish innovation fund Sitra (EW) and Finska Läkaresällskapet. CNS was supported by American Heart Association Postdoctoral Fellowships 15POST24470131 and 17POST33650016. Charles N Rotimi is supported by Z01HG200362. Zhe Wang, Michael H Preuss, and Ruth JF Loos are supported by R01HL142302. NJT is a Wellcome Trust Investiga‑ tor (202802/Z/16/Z), is the PI of the Avon Longitudinal Study of Parents and Children (MRC & WT 217065/Z/19/Z), is supported by the University of Bristol NIHR Biomedical Research Centre (BRC-1215–2001) and the MRC Integrative Epidemiology Unit (MC_UU_00011), and works within the CRUK Integrative Cancer Epidemiology Programme (C18281/ A19169). Ruth E Mitchell is a member of the MRC Integrative Epidemiology Unit at the University of Bristol funded by the MRC (MC_UU_00011/1). Simon Haworth is supported by the UK National Institute for Health Research Academic Clinical Fellowship. Paul S. de Vries was supported by American Heart Association grant number 18CDA34110116. Julia Ramierz acknowledges support by the People Programme of the European Union’s Seventh Framework Programme grant n° 608765 and Marie Sklodowska-Curie grant n° 786833. Maria Sabater-Lleal is supported by a Miguel Servet contract from the ISCIII Spanish Health Institute (CP17/00142) and co-financed by the European Social Fund. Jian Yang is funded by the Westlake Education Foundation. Olga Giannakopoulou has received funding from the British Heart Foundation (BHF) (FS/14/66/3129). CHARGE Consortium cohorts were supported by R01HL105756. Study-specific acknowledgements are available in the Additional file 32: Supplementary Note. The views expressed in this manuscript are those of the authors Kanoni et al. Genome Biology (2022) 23:268 Page 32 of 42 and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services. Availability of data and materials The GWAS meta-analysis results (including both ancestry-specific and trans-ancestry analyses) and risk score weights are available at: http://​csg.​sph.​umich.​edu/​willer/​public/​glgc-​lipid​s2021 [107]. A web browser displaying the gene prioritization and PheWAS results is available at https://​hugea​mp.​org:​8000/​resea​rch.​html?​pageid=​GLGC_​149 [108]. The optimized trans-ancestry polygenic score weights are deposited within the PGS Catalog (https://​www.​pgsca​talog.​org/​ publi​cation/​PGP00​0230/ [109] and https://​www.​pgsca​talog.​org/​publi​cation/​PGP00​0366/ [110]. Scripts used for analysis and summary of results are available under the MIT license on this GitHub repository: https://​github.​com/​Global-​Lipids-​ Genet​ics [111]. The version of source code used in the manuscript is deposited in Zenodo: https://​doi.​org/​10.​5281/​ zenodo.​71302​99 [112]. Declarations Ethics approval and consent to participate The overall study was approved by the IRB of the Boston University Medical Center. Individual studies were approved by the appropriate institutional review boards (IRB) and informed consent was obtained from all participants. Competing interests Ioanna Ntalla is an employee and stock owner of Gilead Sciences since August 2019. Derek Klarin accepts consulting fees from Regeneron Pharmaceuticals. All deCODE-affiliated authors (Gudmar Thorleifsson, Anna Helgadottir, Daniel F Gudb‑ jartsson, Hilma Holm, Unnur Thorsteinsdottir, Kari Stefansson) are employees of deCODE/Amgen Inc. As of January 2020, Anubha Mahajan is an employee of Genentech, and a holder of Roche stock. Markus Scholz receives funding from Pfizer Inc. for a project not related to this research. Marcus E Kleber is employed by SYNLAB MVZ Mannheim GmbH. Gabriel Cuellar-Partida contributed to this work while employed at The University of Queensland, but he is now an employee of 23andMe Inc. Mark J Caulfield is Chief Scientist for Genomics England, a UK Government company. The views expressed in this article are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health. Mark I McCarthy has served on advisory panels for Pfizer, NovoNordisk, and Zoe Global and has received honoraria from Merck, Pfizer, Novo Nordisk, and Eli Lilly and research funding from Abbvie, Astra Zeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Merck, NovoNordisk, Pfizer, Roche, Sanofi Aventis, Servier, and Takeda. As of June 2019, Mark I McCarthy is an employee of Genentech, and a holder of Roche stock. Winfried März has received grants from Siemens Healthineers, grants and personal fees from Aegerion Pharmaceuticals, grants and personal fees from AMGEN, grants from Astrazen‑ eca, grants and personal fees from Sanofi, grants and personal fees from Alexion Pharmaceuticals, grants and personal fees from BASF, grants and personal fees from Abbott Diagnostics, grants and personal fees from Numares AG, grants and personal fees from Berlin-Chemie, grants and personal fees from Akzea Therapeutics, grants from Bayer Vital GmbH, grants from bestbion dx GmbH, grants from Boehringer Ingelheim Pharma GmbH Co KG, grants from Immundiagnostik GmbH, grants from Merck Chemicals GmbH, grants from MSD Sharp and Dohme GmbH, grants from Novartis Pharma GmbH, grants from Olink Proteomics, other from Synlab Holding Deutschland GmbH, all outside the submitted work. Bruce M Psaty serves on the Steering Committee of the Yale Open Data Access Project funded by Johnson & Johnson. Amit V Khera has served as a consultant to Sanofi, Medicines Company, Maze Pharmaceuticals, Navitor Pharmaceuticals, Verve Therapeutics, Amgen, and Color Genomics; received speaking fees from Illumina, the Novartis Institute for Bio‑ medical Research; received sponsored research agreements from the Novartis Institute for Biomedical Research and IBM Research, and reports a patent related to a genetic risk predictor (20190017119). Dennis O Mook-Kanamori is a part-time clinical research consultant for Metabolon, Inc. Danish Saleheen has received support from the British Heart Foundation, Pfizer, Regeneron, Genentech, and Eli Lilly pharmaceuticals. Veikko Salomaa has received honoraria for consultations from Novo Nordisk and Sanofi and has ongoing research collaboration with Bayer Ltd, all unrelated to the present study. Sekar Kathiresan is an employee of Verve Therapeutics, and holds equity in Verve Therapeutics, Maze Therapeutics, Catabasis, and San Therapeutics. He is a member of the scientific advisory boards for Regeneron Genetics Center and Corvidia Therapeutics; he has served as a consultant for Acceleron, Eli Lilly, Novartis, Merck, Novo Nordisk, Novo Ventures, Ionis, Alnylam, Aegerion, Haug Partners, Noble Insights, Leerink Partners, Bayer Healthcare, Illumina, Color Genomics, MedGenome, Quest, and Medscape; he reports patents related to a method of identifying and treating a person having a predisposition to or afflicted with cardiometabolic disease (20180010185) and a genetics risk predictor (20190017119). Cristen J Willer’s spouse is employed by Regeneron. Author details 1 William Harvey Research Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ, UK. 2 Department of Internal Medicine, Division of Cardiology, University of Michigan, Ann Arbor, MI 48109, USA. 3 Department of Biostatistics, Boston University School of Public Health, 801 Massachusetts Ave, Boston, MA 02118, USA. 4 Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA. 5 Department of Statistics, The Pennsylvania State University, University Park, PA, USA. 6 Huck Institutes of the Life Sciences, The Pennsylvania State University, University Park, PA, USA. 7 VA Palo Alto Health Care Systems, Palo Alto, CA, USA. 8 Department of Statistics, Stanford University, Stanford, CA, USA. 9 Department of Medicine, Division of Cardiovascular Medicine, Stanford University School of Medicine, Stanford, CA 94305, USA. 10 Boston Children’s Hospital, EndocrinologyBoston, MA 02115, USA. 11 Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA, USA. 12 Department of Genetic Epidemiology, University of Regens‑ burg, Regensburg, Germany. 13 McDonnell Genome Institute and Department of Medicine, Washington University, St. Louis, MO 63108, USA. 14 Department of Biostatistics, Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA. 15 Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 48109, USA. 16 Clinical Pharmacology, William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK. 17 Department of Epidemiology, Emory University Rollins Kanoni et al. Genome Biology (2022) 23:268 Page 33 of 42 School of Public Health, Atlanta, GA, USA. 18 Atlanta VA Health Care System, Decatur, GA, USA. 19 Department of Surgery, Stanford University School of Medicine, Stanford, CA, USA. 20 deCODE Genetics/Amgen, Inc. Sturlugata 8, Reykjavik 102, Iceland. 21 School of Engineering and Natural Sciences, University of Iceland, Sæmundargötu 2, Reykjavik 102, Iceland. 22 Department of Clinical Biochemistry, Landspitali - National University Hospital of Iceland, Hringbraut, Reykjavik 101, Iceland. 23 Division of Genome Science, Department of Precision Medicine, National Institute of Health, Chungcheong‑ buk‑Do, South Korea. 24 Laboratory for Statistical Analysis, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan. 25 Department of Ophthalmology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. 26 Department of Statistical Genetics, Osaka University Graduate School of Medicine, Osaka, Japan. 27 Department of Allergy and Rheumatology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. 28 Laboratory for Statistical and Translational Genetics, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan. 29 Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA. 30 Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, MA, USA. 31 K.G. Jebsen Center for Genetic Epidemiology, Department of Public Health and Nursing, NTNU, Norwegian University of Science and Technology, Trondheim, Norway. 32 MRC Integrative Epidemiology Unit (IEU), Bristol Medical School, University of Bristol, Oakfield House, Oakfield Grove, Bristol BS8 2BN, UK. 33 Clinic of Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway. 34 Division of Medicine and Laboratory Sciences, University of Oslo, Oslo, Norway. 35 Institute for Molecular Medicine Finland (FIMM), HiLIFE, University of Helsinki, Tukholmankatu 8, 00014 Helsinki, Finland. 36 Department of Public Health and Welfare, Finnish Institute for Health and Welfare, Helsinki, Finland. 37 Department of Genetics, Institute for Biomedical Informatics, University of Pennsylvania, Perelman School of Medicine, Philadelphia, PA 19104, USA. 38 Center for Genetic Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL 60618, USA. 39 Department of Medicine (Medical Genetics), University of Washington, Seattle, WA, USA. 40 Division of Biomedical Informatics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA. 41 Division of Clinical Pharmacology, Department of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA. 42 Department of Cardiovascular Medicine and the Gonda Vascular Center, Mayo Clinic, Rochester, MN, USA. 43 Tohoku Medical Megabank Organization, Tohoku University, Sendai 980‑8573, Japan. 44 Wellcome Trust Sanger Institute, Hinxton CB10 1SA, UK. 45 Blizard Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, UK. 46 Fred Hutchinson Cancer Center, Division of Public Health Sciences, Seattle, WA 9810, USA. 47 Division of Preventive Medicine, Brigham and Women’s Hospital, Boston, MA 02215, USA. 48 Centre for Genomic and Experimental Medicine, Institute of Genetics and Cancer, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, UK. 49 Usher Institute, The University of Edinburgh, Nine, Edinburgh Bioquarter, 9 Little France Road, Edinburgh EH16 4UX, UK. 50 Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, Oxford OX3 7LF, UK. 51 Medical Research Council Population Health Research Unit, Nuffield Department of Population Health, University of Oxford, Oxford OX3 7LF, UK. 52 Center for Non-Communicable Diseases, Karachi, Sindh, Pakistan. 53 Department of Population Medicine, Qatar University College of Medicine, QU Health, Doha, Qatar. 54 Survey Research Center, Institute for Social Research, University of Michigan, Ann Arbor, MI 48104, USA. 55 Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI 48109, USA. 56 Program in Genetic Epidemiology and Statistical Genetics, Department of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA. 57 Department of Epidemiol‑ ogy, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC, USA. 58 Sydney Brenner Institute for Molecular Bioscience, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa. 59 Wellcome Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK. 60 Human Genetics Center, Depart‑ ment of Epidemiology, School of Public Health, Human Genetics, and Environmental Sciences, The University of Texas Health Science Center at Houston, Houston, TX 77030, USA. 61 Department of Epidemiology and Biostatistics, Imperial College London, London W2 1PG, UK. 62 Department of Cardiology, Ealing Hospital, London North West University Healthcare NHS Trust, Middlesex UB1 3HW, UK. 63 Department of Medical Sciences, Molecular Epidemiology and Science for Life Laboratory, Uppsala University, Uppsala, Sweden. 64 MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Cambridge CB2 0QQ, UK. 65 Cardiovascular Epidemiology Unit, Department of Public Health and Primary Care, University of Cambridge, Strangeways Research Laboratory, Wort’s Causeway, Cambridge CB1 8RN, UK. 66 Center for Genomic Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan. 67 Department of Internal Medicine, Erasmus MC, University Medical Center Rotterdam, Rotterdam, the Netherlands. 68 Department of Biological Psychology, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands. 69 Amsterdam Public Health Research Institute, Amsterdam UMC, the Netherlands. 70 Genetics of Complex Traits, University of Exeter Medical School, University of Exeter, Exeter EX2 5DW, UK. 71 Nanjing University of Chinese Medicine, Nanjing 210029, Jiangsu, China. 72 Research Unit of Molecular Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany. 73 Institute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environ‑ mental Health, Neuherberg, Germany. 74 Bristol Dental School, University of Bristol, Lower Maudlin Street, Bristol BS1 2LY, UK. 75 Department of Genetic Medicine, Weill Cornell Medicine-Qatar, Doha, Qatar. 76 Department of Computer and Systems Engineering, Alexandria University, Alexandria, Egypt. 77 Population Health Sciences, Bristol Medical School, University of Bristol, Oakfield Grove, Bristol BS8 2BN, UK. 78 Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore 117549, Singapore. 79 Center for Clinical Research and Prevention, Bispebjerg and Frederiksberg Hospital, Copenhagen, Denmark. 80 Department of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. 81 The Institute for Transla‑ tional Genomics and Population Sciences, Department of Pediatrics, Lundquist Institute for Biomedical Innovations (Formerly LABioMed) at Harbor-UCLA Medical Center, Torrance, CA 90502, USA. 82 Center for Public Health Genomics, University of Virginia, Charlottesville, VA 22903, USA. 83 Section of Endocrinology and Metabolism, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan. 84 Institute of Preventive Medicine, National Defense Medical Center, Postbox 90048~700, Sanhsia Dist, New Taipei City 237101, Taiwan. 85 William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, John Vane Science Centre, Charterhouse Square, London EC1M 6BQ, UK. 86 NIHR Barts Cardiovascular Biomedical Research Centre, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK. 87 Aragon Institute of Engineering Research, University of Zaragoza and Centro de Investigación Biomédica en Red - Bioingeniería, Biomateriales Y Nanomedicina, Spain. 88 Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health Kanoni et al. Genome Biology (2022) 23:268 Page 34 of 42 and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. 89 Division of Cardiovascular Medicine, Radcliffe Department of Medicine, John Radcliffe Hospital, University of Oxford, Oxford OX3 9DU, UK. 90 Unit of Genomics of Complex Diseases, Institut d’Investigació Biomèdica Sant Pau (IIB SANT PAU), Barcelona, Spain. 91 Cardiovascular Medicine Unit, Department of Medicine, Karolinska Institutet, Center for Molecular Medicine, Karolinska University Hospital, Stockholm, Sweden. 92 Department of Internal Medicine, Section of Gerontology and Geriatrics, Leiden University Medical Center, Leiden, the Netherlands. 93 Istituto Di Ricerca Genetica E Biomedica, Consiglio Nazionale Delle Ricerche, Rome, Italy. 94 Dipartimento Di Scienze Biomediche, Università Degli Studi Di Sassari, Sardinia, Italy. 95 University Center for Primary Care and Public Health, University of Lausanne, Rte de Berne 113, 1010 Lausanne, Switzerland. 96 Swiss Institute of Bioinformatics, 1015 Lausanne, Switzerland. 97 Department of Medicine, Internal Medicine, Lausanne University Hospital and University of Lausanne, Rue du Bugnon 46, 1011 Lausanne, Switzerland. 98 Department of Epidemiology and Biostatistics, MRC-PHE Centre for Environment and Health, School of Public Health, Imperial College London, London, UK. 99 Department of Cardiology, Leiden University Medical Center, Leiden, the Netherlands. 100 Dept of Internal Medicine, Section of Gerontology and Geriatrics, Leiden University Medical Center, Leiden, the Netherlands. 101 BHF Glasgow Cardiovascular Research Centre, Faculty of Medicine, Glasgow, UK. 102 Institute for Cardiogenetics, University of Lübeck, DZHK (German Research Centre for Cardiovascular Research), Partner Site Hamburg/Lübeck/Kiel, University Heart Center Lübeck, Lübeck and Charité – University Medicine Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität Zu Berlin and Berlin Institute of Health, Institute for Dental and Craniofacial Sciences, Department of Periodontology and Synoptic Dentistry, Berlin, Germany. 103 Deutsches Herzzentrum München, Klinik Für Herz- Und Kreislauferkrankungen, Technische Universität München, Munich, Germany. 104 Deutsches Zentrum Für Herz‑Kreislauf‑Forschung (DZHK) E.V., Partner Site Munich Heart Alliance, Munich, Germany. 105 Key Laboratory of Cardiovascular Epidemiology and Department of Epidemiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100037, China. 106 Lund University Diabetes Center, Lund University, Malmö, Sweden. 107 Institute of Genetic Epidemiology, Department of Genetics, Medical University of Innsbruck, Innsbruck, Austria. 108 German Chronic Kidney Disease Study, Berlin, Germany. 109 Institute for Medical Informatics, Statistics and Epidemiology, University of Leipzig, Haertelstrasse 16‑18, 04107 Leipzig, Germany. 110 LIFE Research Centre for Civilization Diseases, University of Leipzig, Philipp‑Rosenthal‑Straße 27, 04103 Leipzig, Germany. 111 Radboud University Medical Center, Radboud Institute for Health Sciences, Nijmegen, The Netherlands. 112 Quantinuum Research LLC, Wayne, PA 19087, USA. 113 Division of Statistical Genomics, Department of Genetics, Washington University School of Medicine, St. Louis, MO, USA. 114 Department of Epidemiology, University of Pittsburgh, Pittsburgh, PA, USA. 115 Institute for Biomedicine, Eurac Research, Affiliated Institute of the University of Lübeck, Via Galvani 31, 39100 Bolzano, Italy. 116 Department of Clinical Biochemistry, Lillebaelt Hospital, Vejle, Denmark. 117 Cardiovascular Health Research Unit, Department of Medicine, University of Washington, Seattle 98101, USA. 118 Unidad de Investigacion Medica en Bioquimica, Hospital de Especiali‑ dades, Centro Medico Nacional Siglo XXI, Instituto Mexicano del Seguro Social, Mexico City, Mexico. 119 The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 120 Depart‑ ment of Twin Research and Genetic Epidemiology, King’s College London, London SE1 7EH, UK. 121 NIHR Biomedical Research Centre at Guy’s and St Thomas’ Foundation Trust, London SE1 9RT, UK. 122 Department of Cardiology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 123 Institute of Cardiovascular Sciences, University College London, Gower Street, London WC1E 6BT, UK. 124 Department of Epidemiology and Public Health, University College London, 1‑19 Torrington Place, London WC1E 6BT, UK. 125 Department of Surgery, University of Pennsylvania, Philadelphia, PA, USA. 126 Amsterdam UMC, Department of Epidemiology and Data Science, Amsterdam Public Health Research Institute, Amsterdam 1081HV, the Netherlands. 127 Department of Cell and Chemical Biology, Leiden University Medical Center, Leiden 2333ZA, The Netherlands. 128 Montreal Heart Institute, Université de Montréal, 5000 Belanger Street, Montreal, PQ H1T1C8, Canada. 129 Icelandic Heart Association, 201 Kopavogur, Iceland. 130 Depart‑ ment of Anthropology, University of Toronto at Mississauga, Mississauga, ON L5L 1C6, Canada. 131 Vth Department of Medicine, Medical Faculty Mannheim, Heidelberg University, 68167 Mannheim, Germany. 132 SYNLAB MVZ Humange‑ netik Mannheim GmbH, 68163 Mannheim, Germany. 133 Shanghai Institute of Nutrition and Health, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Shanghai, China. 134 Biomedical Technology Research Center, Tokushima Research Institute, Otsuka Pharmaceutical Co., Ltd, Tokushima, Japan. 135 School of Life Sciences, Westlake University, Hangzhou 310024, Zhejiang, China. 136 Westlake Laboratory of Life Sciences and Biomedicine, Hang‑ zhou 310024, Zhejiang, China. 137 Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD 4072, Australia. 138 Nuffield Department of Population Health, University of Oxford, Oxford, UK. 139 Department of Epidemiology, Erasmus MC, University Medical Center Rotterdam, Rotterdam, the Netherlands. 140 Section of Statistical Multi‑Omics, Department of Clinical and Experimental Research, University of Surrey, Guildford, Surrey, UK. 141 Laboratory of Neuroge‑ netics, National Institute On Aging, NIH, Bethesda, MD, USA. 142 Data Tecnica International, Glen Echo, MD, USA. 143 MRC Human Genetics Unit, Institute of Genetics and Cancer, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, Scotland. 144 Institute for Medical Informatics, Biometry and Epidemiology, University of Duisburg- Essen, Essen, Germany. 145 Centre for Public Health, Queen’s University of Belfast, Belfast, Northern Ireland. 146 Genomic Oncology Area, GENYO, Centre for Genomics and Oncological Research: Pfizer-University of Granada-Andalusian Regional Government, Granada, Spain. 147 Hematology Department, Hospital Universitario Virgen de Las Nieves, Granada, Spain. 148 Instituto de Investigación Biosanitaria de Granada (Ibs.GRANADA), Granada, Spain. 149 Department of Epidemiol‑ ogy and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China. 150 Genome Institute of Singapore, Agency for Science, Technology and Research, Singapore, Singapore. 151 Public Health Informatics Unit, Department of Integrated Health Sciences, Nagoya University Graduate School of Medicine, Nagoya 461‑8673, Japan. 152 MRC Unit for Lifelong Health and Ageing at UCL, 1‑19 Torrington Place, London WC1E 7HB, UK. 153 Tampere Centre for Skills Training and Simulation, Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland. 154 Brown Foundation Institute of Molecular Medicine, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, TX 77030, USA. 155 CONACYT, Instituto Nacional de Ciencias Médicas Y Nutrición Salvador Zubirán, Ciudad de Mexico, Mexico. 156 Programs in Metabolism and Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA. 157 Departamento de Medicina Genómica Y Toxicología Ambiental, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, Kanoni et al. Genome Biology (2022) 23:268 Page 35 of 42 Coyoacán, 04510 Mexico City, Mexico. 158 Departamento de Genómica Computacional, Instituto Nacional de Medicina Genómica, Mexico City, Mexico. 159 Center for Diabetes Research, University of Bergen, Bergen, Norway. 160 Genomic Research On Complex Diseases (GRC Group), CSIR-Centre for Cellular and Molecular Biology, Hyderabad, Telangana, India. 161 Academy of Scientific and Innovative Research (AcSIR), New Delhi, India. 162 Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA. 163 Hunter Medical Research Institute, Newcastle, Australia. 164 Medical Research Institute, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul 03181, Korea. 165 Department of Clinical Research Design & Evaluation, SAIHST, Sungkyunkwan University, Seoul 06355, Korea. 166 Center for Cohort Studies, Total Healthcare Center, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul 04514, Korea. 167 Department of Occupational and Environmental Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul 03181, Korea. 168 Centre for Global Health Research, Usher Institute, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, Scotland. 169 Thurston Arthritis Research Center, University of North Carolina, Chapel Hill, NC, USA. 170 Duke‑NUS Medical School, Health Services and Systems Research, Singapore 169857, Singapore. 171 Division of Biomedical Informatics and Personalized Medicine, Department of Medicine, Anschutz Medical Campus, University of Colorado, Denver, Aurora, CO 80045, USA. 172 Genom‑ ics and Molecular Medicine Unit, CSIR-Institute of Genomics and Integrative Biology, New Delhi 110020, India. 173 Academy of Scientific and Innovative Research, CSIR-Human Resource Development Centre, Ghaziabad, Uttar Pradesh, India. 174 Departments of Ophthalmology and Human Genetics, Radboud University Nijmegen Medical Center, Philips Van Leydenlaan 15, Nijmegen 6525 EX, the Netherlands. 175 Vanderbilt Epidemiology Center, Division of Epidemi‑ ology, Vanderbilt University Medical Center, Nashville, USA. 176 Department of Pediatrics, University of California San Francisco, Oakland, CA 94609, USA. 177 National Center for Global Health and Medicine, Tokyo 1628655, Japan. 178 Department of Genetics, University of North Carolina, Chapel Hill, NC 27599, USA. 179 Department of Biostatistics and Epidemiology, University of Massachusetts-Amherst, Amherst, MA 01003, USA. 180 Department of Cardiovascular Sciences, University of Leicester, Leicester, UK. 181 NIHR Leicester Biomedical Research Centre, Glenfield Hospital, Leicester, UK. 182 Beijing Institute of Ophthalmology, Beijing Key Laboratory of Ophthalmology and Visual Sciences, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, 17 Hougou Lane, Chong Wen Men, Beijing 100005, China. 183 Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, 1 Dong Jiao Min Xiang, Beijing 100730, Dong Cheng District, China. 184 Institute of Genetics and Biophysics “Adriano Buzzati-Traverso” - CNR, Naples, Italy. 185 IRCCS Neuromed, Pozzilli, Isernia, Italy. 186 Division of Biostatistics, Washington University, St. Louis, MO 63110, USA. 187 Rush Alzheimer’s Disease Center, Rush University Medical Center, Chicago, IL, USA. 188 Department of Neurological Sciences, Rush University Medical Center, Chicago, IL, USA. 189 Dept of Nephrology, University Hospital Regensburg, Regensburg, Germany. 190 Institute for Maternal and Child Health, IRCCS Burlo Garofolo, Trieste, Italy. 191 Department of Molecular Epidemiology, German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Germany. 192 German Center for Diabetes Research (DZD), Munich‑Neuherberg, Germany. 193 Department of Genetics and Bioinformatics, Dasman Diabetes Institute, Kuwait City, Kuwait. 194 Department of Nutrition and Dietetics, School of Health Science and Education, Harokopio University of Athens, Athens, Greece. 195 Department of Clinical Epidemiol‑ ogy, Institute of Health Informatics, University College London, London, UK. 196 Clinical Division of Neurogeriatrics, Department of Neurology, Medical University of Graz, Graz, Austria. 197 Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Graz, Austria. 198 Massachusetts General Hospital Cancer Center, Charlestown, MA 02129, USA. 199 Institute of Genetic and Biomedical Research, National Research Council of Italy, UOS of Sassari, Sassari, Italy. 200 Department of Epidemiology, University of Groningen, University Medical Center Groningen, Groningen 9700 RB, the Netherlands. 201 Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku, Turku, Finland. 202 Centre for Population Health Research, University of Turku and Turku University Hospital, Turku, Finland. 203 Sleep Medicine and Circadian Disorders, Brigham and Women’s Hospital, Boston, MA 02115, USA. 204 Division of Sleep Medicine, Harvard Medical School, Boston, MA 02115, USA. 205 Central Diagnostics Laboratory, Division Laboratories, Pharmacy, and Biomedical Genetics, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands. 206 Laboratory of Epidemiology and Population Sciences, National Institute On Aging, NIH, Baltimore, MD 20892‑9205, USA. 207 Department of Engineering Technology, University of Houston-Sugarland, Houston, TX, USA. 208 Interfaculty Institute for Genetics and Functional Genomics, Department of Functional Genomics, University of Greifswald and University Medicine Greifswald, Greifswald, Germany. 209 Center for Research On Genomics and Global Health, National Human Genome Research Institute, National Institutes of Health, 12 South Drive, Room 1025, Bethesda, MD 20892, USA. 210 Oneomics. Co. Ltd. 2F, Soonchunhyang Mirai Medical Center 173, Buheuyng‑Ro, Bucheon‑Si Gyeonggi‑Do 14585, Korea. 211 Department of Clinical Biochemistry and Immunology, Hospital of Southern Jutland, Kresten Philipsens Vej 15, 6200 Aabenraa, Denmark. 212 Department of Clinical Biochemistry, Lillebaelt Hospital, Kolding, Denmark. 213 Department of Biomedical Science, Hallym University, Chuncheon 24252, Gangwon‑Do, Korea. 214 Centre for Bone and Arthritis Research, Department of Internal Medicine and Clinical Nutrition, Institute of Medicine, Sahlgren‑ ska Academy, University of Gothenburg, Gothenburg, Sweden. 215 Bioinformatics Core Facility, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 216 Institute of Medical Informatics and Statistics, Kiel University, Kiel, Germany. 217 Institute of Translational Genomics, Helmholtz Zentrum München – German Research Center for Environ‑ mental Health, Neuherberg, Germany. 218 Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, Oxford, UK. 219 School of Medicine and Public Health, College of Health, Medicine and Wellbeing, University of Newcastle, Newcastle, NSW 2308, Australia. 220 Center for Geriatrics and Gerontology, Division of Endocrinology and Metabolism, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan. 221 School of Medicine, National Yang-Ming University, Taipei, Taiwan. 222 School of Medicine, National Defense Medical Center, Taipei, Taiwan. 223 Division of Endocrinology and Metabolism, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan. 224 Department of Medicine, School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan. 225 Department of Kinesiology, Université Laval, Québec, Canada. 226 Department of Clinical Chemistry, Fimlab Laboratories, 33520 Tampere, Finland. 227 Department of Clinical Chemistry, Finnish Cardiovascular Research Center - Tampere, Faculty of Medicine and Health Technology, Tampere University, 33014 Tampere, Finland. 228 Department of Cardiology, Heart Center, Tampere University Hospital, 33521 Tampere, Finland. 229 Department of Cardiology, Finnish Cardiovascular Research Center - Tampere, Faculty of Medicine and Health Technology, Tampere University, 33014 Tampere, Finland. 230 University of Queensland Diamantina Institute, Translational Research Institute, Kanoni et al. Genome Biology (2022) 23:268 Page 36 of 42 Kent St, Woolloongabba, Brisbane, QLD 4102, Australia. 231 Department of Medicine, Bornholms Hospital, Rønne, Denmark. 232 Department of Epidemiology, Ryals School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA. 233 Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands. 234 Department of Population and Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH 44106, USA. 235 Department of Genetics, Stanford University School of Medicine Stanford, Palo Alto, CA 94305, USA. 236 Division of Endocrinology, Ohio State University, Columbus, OH 43210, USA. 237 Department of Epidemiology, University of Washington, Seattle, WA 98195, USA. 238 School of Medicine and Health Sciences, George Washington University, Washington, DC 20037, USA. 239 Department of Epidemiology, School of Public Health, Peking University Health Science Center, Beijing, China. 240 Institute for Laboratory Medicine, University Hospital Leipzig, Paul‑List‑Strasse 13/15, 04103 Leipzig, Germany. 241 Laboratory of Epidemiology and Population Sciences, National Institute On Aging, NIH, Baltimore, MD 20892‑9205, USA. 242 Center for Alzheimer’s and Related Dementias, NIH, Bethesda, MD, USA. 243 Data Tecnica International, Washington, DC, USA. 244 Department of Clinical Physiology and Nuclear Medicine, Turku University Hospital, Turku, Finland. 245 Department of Environmental and Preventive Medicine, Jichi Medical University School of Medicine, Shimotsuke 329‑0498, Japan. 246 Centre for Population Health Sciences, Usher Institute, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, Scotland. 247 Department of Functional Pathology, Shimane University School of Medicine, Izumo 6938501, Japan. 248 Department of Pediatrics and Adolescent Medicine, Turku University Hospital and University of Turku, Turku, Finland. 249 Department of Physiology, University of Turku, Turku, Finland. 250 University of Split School of Medicine, Šoltanska 2, HR‑21000 Split, Croatia. 251 University of Leipzig Medical Center, Liebigstr. 18, 04103 Medical Department III – Endocrinology, Nephrology, RheumatologyLeipzig, Germany. 252 Department of Nutrition‑Dietetics, Harokopio University, Eleftheriou Venizelou, 17676 Athens, Greece. 253 Department of Clinical Gene Therapy, Osaka University Graduate School of Medicine, Suita 5650871, Japan. 254 Department of Geriatric and General Medicine, Osaka University Graduate School of Medicine, Suita 5650871, Japan. 255 Department of Vascular Surgery, Division of Surgical Specialties, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands. 256 Corneal Dystrophy Research Institute, Yonsei University College of Medicine, Saevit Eye Hospital, SeoulIlsan 03722, Korea. 257 Dept of Radiology and Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Rotterdam, the Netherlands. 258 Latin American Brain Health (BrainLat), Universidad Adolfo Ibáñez, Santiago, Chile. 259 Julius Centre for Health Sciences and Primary Care, University Medical Centre Utrecht, 3584CG Utrecht, the Netherlands. 260 Second Department of Cardiology, Medical School, National and Kapodistrian University of Athens, Attikon University Hospital, Athens, Greece. 261 Center for Vision Research, Department of Ophthalmology and The Westmead Institute, University of Sydney, Hawkesbury Rd, Sydney, NSW 2145, Australia. 262 School of Medicine, Menzies Institute for Medical Research, University of Tasmania, Liverpool St, Hobart, TAS 7000, Australia. 263 Centre for Eye Research Australia, University of Melbourne, Melbourne, VIC 3002, Australia. 264 Department of Clinical Physiology, Tampere University Hospital, 33521 Tampere, Finland. 265 Department of Clinical Physiology, Finnish Cardiovascular Research Center ‑ Tampere, Faculty of Medicine and Health Technology, Tampere University, 33014 Tampere, Finland. 266 Centre Nutrition, Santé Et Société (NUTRISS), Institute of Nutrition and Functional Foods (INAF), Québec, Canada. 267 Pennington Biomedical Research Center, Baton Rouge, LA 70808, USA. 268 Discipline of Internal Medicine, Medical School, The University of Western Australia, Perth, WA, Australia. 269 Institute of Epidemiology, Kiel University, Kiel, Germany. 270 Institute of Clinical Molecular Biology, Kiel University, Kiel, Germany. 271 Department of Drug Treatment, Sahlgrenska University Hospital, Gothenburg, Sweden. 272 Geriatric Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 273 Department of Internal Medicine, EwhaWomans University School of Medicine, Seoul, Korea. 274 Division of Cancer Control and Population Sciences, UPMC Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15232, USA. 275 Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA 15232, USA. 276 Healthy Longevity Translational Research Programme, Yong Loo Lin School of Medicine, National University of Singapore, Singapore 117545, Singapore. 277 Singapore Institute for Clinical Sciences, Agency for Science Technology and Research (A*STAR), Singapore 117609, Singapore. 278 Department of Endocrinology and Metabolism, Kyung Hee University School of Medicine, Seoul 02447, Korea. 279 Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany. 280 Laboratory of Epidemiology and Population Science, National Institute On Aging Intramural Research Program, NIH Biomedical Research Center, NIH 251 Bayview Blvd, Baltimore, MD 21224, USA. 281 Algebra University College, Ilica 242, Zagreb, Croatia. 282 Department of Physical Activity and Health, Paavo Nurmi Centre, Sports and Exercise Medicine Unit, University of Turku, Turku, Finland. 283 Interdisciplinary Center Psychopathology and Emotion Regulation (ICPE), University of Groningen, University Medical Center Groningen, Groningen 9700 RB, the Netherlands. 284 Institute of Molecular Genetics, National Research Council of Italy, Pavia, Italy. 285 Gottfried Schatz Research Center for Cell Signaling, Metabolism and Aging, Medical University of Graz, Graz, Austria. 286 Local Health Unit Toscana Centro, Florence, Italy. 287 Institute of Nutritional Science, University of Potsdam, Nuthetal, Germany. 288 Department of Medicine, Surgery and Health Sciences, University of Trieste, Strada Di Fiume 447, 34149 Trieste, Italy. 289 Department of Nephrology, , Traunstein Hospital, Diabetology, RheumatologyTraunstein, Germany. 290 KfH Kidney Center Traunstein, Traunstein, Germany. 291 Department of Neurology, Center for Translational and Systems Neuroimmunology, Columbia University Medical Center, New York, NY, USA. 292 Medical School, National and Kapodistrian University Athens, 75 M. Assias Street, 115 27 Athens, Greece. 293 Dromokaiteio Psychiatric Hospital, 124 61 Athens, Greece. 294 Clinical Pharmacology, William Harvey Research Institute, Queen Mary University of London, London EC1M 6BQ, UK. 295 Department of Ophthalmology, Medical Faculty Mannheim, Heidelberg University, Kutzerufer 1, 68167 Mannheim, Germany. 296 Institute of Molecular and Clinical Ophthalmology, Basel, Switzerland. 297 Privatpraxis Prof Jonas Und Dr Panda-Jonas, Heidelberg, Germany. 298 Department of Human Genetics, David Geffen School of Medicine at UCLA, University of California, Los Angeles, CA, USA. 299 Unidad de Biología Molecular Y Medicina Genómica, Instituto de Investigaciones Biomédicas UNAM/ Instituto Nacional de Ciencias Médicas Y Nutrición Salvador Zubirán, Mexico City, Mexico. 300 Departamento de Endocrinología Y Metabolismo, Instituto Nacional de Ciencias Médicas Y Nutrición Salvador Zubirán, 14080 Mexico, Mexico. 301 Depart‑ ment of Nutrition, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC 27599, USA. 302 Carolina Population Center, University of North Carolina, Chapel Hill, NC 27516, USA. 303 USC–Office of Population Studies Foundation, University of San Carlos, 6000 Cebu City, Philippines. 304 Department of Anthropology, Sociology, and History, University of San Carlos, 6000 Cebu City, Philippines. 305 Department of Medicine, Faculty of Medicine, University of Kelaniya, Ragama 11010, Sri Lanka. 306 Department of Public Health, Faculty of Medicine, University Kanoni et al. Genome Biology (2022) 23:268 Page 37 of 42 of Kelaniya, Ragama 11010, Sri Lanka. 307 Children’s Hospital Oakland Research Institute, Oakland, CA 94609, USA. 308 Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan. 309 Systems Genomics Laboratory, School of Biotech‑ nology, Jawaharlal Nehru University, New Delhi 110067, India. 310 Department of Medicine, University of Mississippi Medical Center, Jackson, MS 39216, USA. 311 Department of Physiology and Biophysics, University of Mississippi Medical Center, Jackson, MS 39216, USA. 312 Department of Medical Sciences, Uppsala University, Uppsala, Sweden. 313 Depart‑ ment of Paediatrics, Yong Loo Lin School of Medicine, National University of Singapore and Khoo Teck Puat - National University Children’s Medical Institute, National University Health System, Singapore, Singapore. 314 Department of Medicine, University of North Carolina, Chapel Hill, NC, USA. 315 Injury Prevention Research Center, University of North Carolina, Chapel Hill, NC, USA. 316 Division of Physical Therapy, University of North Carolina, Chapel Hill, NC, USA. 317 Department of Psychiatry, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands. 318 Depart‑ ment of Biochemistry, College of Medicine, Ewha Womans University, Seoul 07804, Korea. 319 Faculty of Health and Medicine, University of Newcastle, Callaghan, Australia. 320 Division of Biostatistics, Washington University School of Medicine, St. Louis, USA. 321 Office of the Provost, University of South Carolina, Columbia, SC, USA. 322 Department of Internal Medicine, University of Utah, Salt Lake City, Utah 84132, USA. 323 Institute of Cellular Medicine (Diabetes), The Medical School, Newcastle University, Framlington Place, Newcastle Upon Tyne NE2 4HH, UK. 324 Department of Medi‑ cine, Helsinki University Hospital, University of Helsinki, Haartmaninkatu 4, P.O.Box 340, 00029 Helsinki, Finland. 325 Minerva Foundation Institute for Medical Research, Biomedicum 2U, Tukholmankatu 8, 00290 Helsinki, Finland. 326 JSS Academy of Higher Education and Research, Mysuru, India. 327 Diabetes Unit and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA, USA. 328 Harvard Medical School, Boston, MA 02115, USA. 329 InstitutoNa‑ cional de Salud Publica Y Centro de Estudios en Diabetes, Cuernavaca, Mexico. 330 Laboratorio de Inmunogenómica Y Enfermedades Metabólicas, Instituto Nacional de Medicina Genómica, Mexico City, Mexico. 331 Human Genetics Center, School of Public Health, University of Texas Health Science Center at Houston, Houston, TX 77030, USA. 332 Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore 119228, Singapore. 333 Department of Exercise and Nutrition Sciences, Milken Institute School of Public Health, George Washington University, Washington, DC 20052, USA. 334 Kurume University School of Medicine, Kurume 830‑0011, Japan. 335 Genetics, Merck Sharp & Dohme Corp., Kenilworth, NJ 07033, USA. 336 Population Health and Genomics, University of Dundee, Ninwells Hospital and Medical School, Dundee DD1 9SY, UK. 337 Intramural Research Program, National Institute On Aging, 3001 S. Hanover St., Baltimore, MD 21225, USA. 338 Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing Ophthalmology and Visual Sciences Key Laboratory, Beijing 100730, China. 339 The Eye Hospital, School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou 325027, Zhejiang, China. 340 Synlab Academy, SYNLAB Holding Deutschland GmbH, Mannheim and Augsburg, Germany. 341 Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz, Austria. 342 Faculty of Medicine, University of Iceland, 101 Reykjavik, Iceland. 343 Department of Biomedical Data Sciences, Section Molecular Epidemiology, Leiden University Medical Center, Leiden 2333ZA, The Netherlands. 344 Department of Epidemiology and Data Science, Amsterdam UMC, Amsterdam 1081HV, the Netherlands. 345 Amster‑ dam Public Health Research Institute, Amsterdam Cardiovascular Sciences, Amsterdam 1081HV, the Netherlands. 346 Department of General Practice, Amsterdam UMC, Amsterdam 1081HV, the Netherlands. 347 Amsterdam Public Health Research Institute, Health Behaviours and Chronic Diseases, Amsterdam 1081HV, the Netherlands. 348 Corporal Michael Crescenz VA Medical Center, Philadelphia, PA 19104, USA. 349 Institute of Social and Economic Research, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK. 350 Department of Environmental Medicine and Public Health, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 351 Department of Epidemiology, University of Washington, Seattle, WA, USA. 352 Department of Health Systems and Population Health, University of Washington, Seattle, WA, USA. 353 Institute of Regional Health Research, University of Southern Denmark, Odense, Denmark. 354 Danish Aging Research Center, University of Southern Denmark, Odense C, Denmark. 355 Public Health, Faculty of Medicine, University of Helsinki, Helsinki, Finland. 356 Broad Institute of MIT and Harvard, Cambridge, MA, USA. 357 Center for Applied Genomics, Children’s Hospital of Philadelphia, Philadelphia, PA 19104, USA. 358 Department of Pediatrics, The University of Pennsylvania Perelman School of Medicine, Philadelphia, PA 19104, USA. 359 Division of Human Genetics, Children’s Hospital of Philadelphia, Philadelphia, PA 19104, USA. 360 School of Medicine, Southern University of Science and Technology, Shenzhen, China. 361 Institute for Cardiogenetics, University of Lübeck, DZHK (German Research Centre for Cardiovascular Research), Partner Site Hamburg/Lübeck/Kiel, and University Heart Center Lübeck, Lübeck, Germany. 362 Netherlands Heart Institute, Utrecht, the Netherlands. 363 Division of Cardiology, Department of Medicine, Massachusetts General Hospital, Boston, MA, USA. 364 Department of Medicine, Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA, USA. 365 Department of Medicine, Harvard Medical School, Boston, MA, USA. 366 Verve Therapeutics, Cambridge, MA, USA. 367 Northern Finland Birth Cohorts, Infrastructure for Population Studies, Faculty of Medicine, University of Oulu, Oulu, Finland. 368 Center for Life Course Health Research, Faculty of Medicine, University of Oulu, Oulu, Finland. 369 Biocenter of Oulu, University of Oulu, Oulu, Finland. 370 Institute for Genetic and Biomedical Research, Italian National Council of Research (IRGB CNR), Cagliari, Italy. 371 University of Sassari, Sassari, Italy. 372 Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, the Netherlands. 373 Department of Public Health and Primary Care, Leiden University Medical Center, Leiden, the Netherlands. 374 Department of Internal Medicine, Division of Endocrinology, Leiden University Medical Center, Leiden, the Netherlands. 375 Einthoven Laboratory for Experimental Vascular Medicine, Leiden University Medical Center, Leiden, the Netherlands. 376 Department of Human Genetics, Leiden University Medical Center, Leiden, the Netherlands. 377 Population Health Research Institute, St George’s, University of London, London SW17 0RE, UK. 378 National Heart and Lung Institute, Imperial College London, London W12 0NN, UK. 379 School of Public Health, Imperial College London, London W12 7RH, UK. 380 Department of Internal Medicine, National Taiwan University Hospital, No. 7, Chung‑Shan South Road, Taipei, Taiwan. 381 OCDEM, University of Oxford, Churchill Hospital, Oxford OX3 7LE, UK. 382 NIHR Oxford Biomedical Research Centre, Churchill Hospital, Oxford, UK. 383 Ocular Epidemiology, Singapore Eye Research Institute, Singapore National Eye Centre, Singapore 168751, Singapore. 384 Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singa‑ pore 169857, Singapore. 385 Data Science, Singapore Eye Research Institute, Singapore National Eye Centre, Singa‑ pore 168751, Singapore. 386 Medical School, University of Exeter, University of Exeter, Exeter EX2 5DW, UK. 387 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. 388 Framingham Heart Study, Kanoni et al. Genome Biology (2022) 23:268 Page 38 of 42 National Heart, Lung, and Blood Institute, US National Institutes of Health, Bethesda, MD, USA. 389 Department of Genetics, School of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran. 390 Department of Genetics, Shanghai‑MOST Key Laboratory of Health and Disease Genomics, Chinese National Human Genome Center at Shanghai, Shanghai 201203, China. 391 Computational Medicine, Berlin Institute of Health at Charité – Universitätsmedizin, Berlin, Germany. 392 Technical University of Munich (TUM) and Klinikum Rechts Der Isar, TUM School of Medicine, Munich, Germany. 393 Institute of Clinical Medicine, Internal Medicine, University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland. 394 Stanford Cardiovascular Institute, Stanford University, Stanford, CA 94305, USA. 395 Stanford Diabetes Research Center, Stanford University, Stanford, CA 94305, USA. 396 Regeneron Pharmaceuticals, Tarrytown, NY, USA. 397 Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore 308232, Singapore. 398 Imperial College Healthcare NHS Trust, Imperial College London, London W12 0HS, UK. 399 Imperial College Healthcare NHS Trust, London W12 0HS, UK. 400 MRC-PHE Centre for Environment and Health, Imperial College London, London W2 1PG, UK. 401 School of Electrical and Information Engineering, University of the Witwatersrand, Johannesburg, South Africa. 402 Institute for Minority Health Research, University of Illinois College of Medicine, Chicago, IL, USA. 403 Department of Biostatistics, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA. 404 QIMR Berghofer Medical Research Institute, 300 Herston Road, Brisbane, QLD 4006, Australia. 405 Faisalabad Institute of Cardiol‑ ogy, Faislabad, Pakistan. 406 Department of Medicine, Columbia University Irving Medical Center, New York, NY, USA. 407 Department of Cardiology, Columbia University Irving Medical Center, New York, NY, USA. 408 Big Data Institute, University of Oxford, Oxford OX3 7LF, UK. 409 National Institute for Health Research Oxford Biomedical Research Centre, Oxford University Hospitals, Oxford, UK. 410 Aberdeen Centre for Health Data Science,1:042 Polwarth Building School of Medicine, Medical Science and Nutrition University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK. 411 Division of Population Health and Genomics, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, UK. 412 Department of Population Health Sciences, Geisinger Health, Danville, PA 17822, USA. 413 School of Basic and Medical Biosciences, Faculty of Life Sciences and Medicine, King’s College London, London, UK. 414 Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA. 415 Departments of Medicine (Medical Genetics) and Genome Sciences, University of Washington, Washington, USA. 416 Center for Autoimmune Genomics and Etiology, Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH, USA. 417 Division of Endocrinology, Metabolism, and Molecular Medicine, Department of Medicine, Northwestern University, Feinberg School of Medicine, Chicago, IL 60618, USA. 418 Department of Anthropology, Northwestern University, Evanston, IL 60208, USA. 419 Department of Public Health and Nursing, HUNT Research Centre, NTNU, Norwegian University of Science and Technology, 7600 Levanger, Norway. 420 Department of Research, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway. 421 Depart‑ ment of Endocrinology, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway. 422 RIKEN Center for Integrative Medical Sciences, Kanagawa, Japan. 423 Laboratory for Statistical Analysis, RIKEN Center for Integrative Medical Sciences, Kanagawa, Japan. 424 Laboratory of Complex Trait Genomics, Department of Computational Biology and Medical Sciences, Graduate School of Frontier Sciences, The University of Tokyo, Tokyo, Japan. 425 Laboratory for Systems Genetics, RIKEN Center for Integrative Medical Sciences, Kanagawa, Japan. 426 Department of Genome Informatics, Graduate School of Medicine, the University of Tokyo, Tokyo, Japan. 427 Division of Molecular Pathology, Institute of Medical Science, The University of Tokyo, Tokyo, Japan. 428 Faculty of Medicine, University of Iceland, Sæmundargötu 2, Reykjavik 102, Iceland. 429 VA Boston Healthcare System, Boston, MA, USA. 430 VA Informatics and Computing Infrastructure, VA Salt Lake City Health Care System, Salt Lake City, UT, USA. 431 University of Massachu‑ setts, Boston, MA, USA. 432 Cardiovascular Institute, Stanford University School of Medicine, Stanford, CA, USA. 433 Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA. 434 Department of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. 435 Department of Medicine, Brigham Women’s Hospital, Boston, MA, USA. 436 Division of Cardiology, Emory University School of Medicine, Atlanta, GA, USA. 437 Departments of Pediatrics and Genetics, Harvard Medical School, Boston, MA, USA. 438 Centre for Genetics and Genomics Versus Arthritis, Centre for Musculoskeletal Research, Division of Musculoskeletal and Dermatological Sciences, The University of Manchester, Manchester, UK. 439 Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. 440 Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. 441 Cardiovascular Research Center and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA, USA. 442 Princess Al-Jawhara Al-Brahim Centre of Excellence in Research of Hereditary Disorders (PACER-HD), King Abdulaziz University, Jeddah, Saudi Arabia. 443 Department of Human Genetics, University of Michigan, Ann Arbor, MI 48109, USA. Received: 5 April 2022 Accepted: 6 December 2022 References 1. Castelli WP, Anderson K, Wilson PW, Levy D. Lipids and risk of coronary heart disease. The Framingham Study. Ann Epidemiol. 1992;2:23–8. 2. GBD. Diseases and Injuries Collaborators: Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2019;2020(396):1204–22. 3. Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, Braun LT, de Ferranti S, Faiella-Tommasino J, Forman DE, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2019;139:e1082–143. 4. Diabetes Genetics Initiative of Broad Institute of Harvard and MIT, Lund University, and Novartis Institutes of BioMedical Research, Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, Chen H, Roix JJ, et al. Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science. 2007;316:1331–6. Kanoni et al. Genome Biology (2022) 23:268 Page 39 of 42 5. Kathiresan S, Manning AK, Demissie S, D’Agostino RB, Surti A, Guiducci C, Gianniny L, Burtt NP, Melander O, Orho- Melander M, et al. A genome-wide association study for blood lipid phenotypes in the Framingham Heart Study. BMC Med Genet. 2007;8(Suppl 1):S17. 6. Kathiresan S, Melander O, Anevski D, Guiducci C, Burtt NP, Roos C, Hirschhorn JN, Berglund G, Hedblad B, Groop L, et al. Polymorphisms associated with cholesterol and risk of cardiovascular events. N Engl J Med. 2008;358:1240–9. 7. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, Koseki M, Pirruccello JP, Ripatti S, Chasman DI, Willer CJ, et al. Biological, clinical and population relevance of 95 loci for blood lipids. Nature. 2010;466:707–13. 8. Asselbergs FW, Guo Y, van Iperen EP, Sivapalaratnam S, Tragante V, Lanktree MB, Lange LA, Almoguera B, Appel‑ man YE, Barnard J, et al. Large-scale gene-centric meta-analysis across 32 studies identifies multiple lipid loci. Am J Hum Genet. 2012;91:823–38. 9. Albrechtsen A, Grarup N, Li Y, Sparso T, Tian G, Cao H, Jiang T, Kim SY, Korneliussen T, Li Q, et al. Exome sequencing- driven discovery of coding polymorphisms associated with common metabolic phenotypes. Diabetologia. 2013;56:298–310. 10. Tachmazidou I, Dedoussis G, Southam L, Farmaki AE, Ritchie GR, Xifara DK, Matchan A, Hatzikotoulas K, Rayner NW, Chen Y, et al. A rare functional cardioprotective APOC3 variant has risen in frequency in distinct population isolates. Nat Commun. 2013;4:2872. 11. Willer CJ, Schmidt EM, Sengupta S, Peloso GM, Gustafsson S, Kanoni S, Ganna A, Chen J, Buchkovich ML, Mora S, et al. Discovery and refinement of loci associated with lipid levels. Nat Genet. 2013;45:1274–83. 12. Holmen OL, Zhang H, Fan Y, Hovelson DH, Schmidt EM, Zhou W, Guo Y, Zhang J, Langhammer A, Lochen ML, et al. Systematic evaluation of coding variation identifies a candidate causal variant in TM6SF2 influencing total choles‑ terol and myocardial infarction risk. Nat Genet. 2014;46:345–51. 13. Peloso GM, Auer PL, Bis JC, Voorman A, Morrison AC, Stitziel NO, Brody JA, Khetarpal SA, Crosby JR, Fornage M, et al. Association of low-frequency and rare coding-sequence variants with blood lipids and coronary heart disease in 56,000 whites and blacks. Am J Hum Genet. 2014;94:223–32. 14. Surakka I, Horikoshi M, Magi R, Sarin AP, Mahajan A, Lagou V, Marullo L, Ferreira T, Miraglio B, Timonen S, et al. The impact of low-frequency and rare variants on lipid levels. Nat Genet. 2015;47:589–97. 15. Tang CS, Zhang H, Cheung CY, Xu M, Ho JC, Zhou W, Cherny SS, Zhang Y, Holmen O, Au KW, et al. Exome-wide association analysis reveals novel coding sequence variants associated with lipid traits in Chinese. Nat Commun. 2015;6:10206. 16. van Leeuwen EM, Karssen LC, Deelen J, Isaacs A, Medina-Gomez C, Mbarek H, Kanterakis A, Trompet S, Postmus I, Verweij N, et al. Genome of The Netherlands population-specific imputations identify an ABCA6 variant associated with cholesterol levels. Nat Commun. 2015;6:6065. 17. Iotchkova V, Huang J, Morris JA, Jain D, Barbieri C, Walter K, Min JL, Chen L, Astle W, Cocca M, et al. Discovery and refinement of genetic loci associated with cardiometabolic risk using dense imputation maps. Nat Genet. 2016;48:1303–12. 18. Liu DJ, Peloso GM, Yu H, Butterworth AS, Wang X, Mahajan A, Saleheen D, Emdin C, Alam D, Alves AC, et al. Exome- wide association study of plasma lipids in >300,000 individuals. Nat Genet. 2017;49:1758–66. 19. Lu X, Peloso GM, Liu DJ, Wu Y, Zhang H, Zhou W, Li J, Tang CS, Dorajoo R, Li H, et al. Exome chip meta-analysis iden‑ tifies novel loci and East Asian-specific coding variants that contribute to lipid levels and coronary artery disease. Nat Genet. 2017;49:1722–30. 20. Hoffmann TJ, Theusch E, Haldar T, Ranatunga DK, Jorgenson E, Medina MW, Kvale MN, Kwok PY, Schaefer C, Krauss RM, et al. A large electronic-health-record-based genome-wide study of serum lipids. Nat Genet. 2018;50:401–13. 21. Kanai M, Akiyama M, Takahashi A, Matoba N, Momozawa Y, Ikeda M, Iwata N, Ikegawa S, Hirata M, Matsuda K, et al. Genetic analysis of quantitative traits in the Japanese population links cell types to complex human diseases. Nat Genet. 2018;50:390–400. 22. Klarin D, Damrauer SM, Cho K, Sun YV, Teslovich TM, Honerlaw J, Gagnon DR, DuVall SL, Li J, Peloso GM, et al. Genetics of blood lipids among ~300,000 multi-ethnic participants of the Million Veteran Program. Nat Genet. 2018;50:1514–23. 23. Spracklen CN, Chen P, Kim YJ, Wang X, Cai H, Li S, Long J, Wu Y, Wang YX, Takeuchi F. Association analyses of East Asian individuals and trans-ancestry analyses with European individuals reveal new loci associated with choles‑ terol and triglyceride levels. Hum Mol Genet. 2018;27:1122. 24. Graham SE, Clarke SL, Wu KH, Kanoni S, Zajac GJM, Ramdas S, Surakka I, Ntalla I, Vedantam S, Winkler TW, et al. The power of genetic diversity in genome-wide association studies of lipids. Nature. 2021;600:675–9. 25. Musunuru K, Strong A, Frank-Kamenetsky M, Lee NE, Ahfeldt T, Sachs KV, Li X, Li H, Kuperwasser N, Ruda VM, et al. From noncoding variant to phenotype via SORT1 at the 1p13 cholesterol locus. Nature. 2010;466:714–9. 26. Musunuru K, Pirruccello JP, Do R, Peloso GM, Guiducci C, Sougnez C, Garimella KV, Fisher S, Abreu J, Barry AJ, et al. Exome sequencing, ANGPTL3 mutations, and familial combined hypolipidemia. N Engl J Med. 2010;363:2220–7. 27. Pers TH, Karjalainen JM, Chan Y, Westra HJ, Wood AR, Yang J, Lui JC, Vedantam S, Gustafsson S, Esko T, et al. Biologi‑ cal interpretation of genome-wide association studies using predicted gene functions. Nat Commun. 2015;6:5890. 28. Weeks EM, Ulirsch JC, Cheng NY, Trippe BL, Fine RS, Miao J, Patwardhan TA, Kanai M, Nasser J, Fulco CP, et al: Lever‑ aging polygenic enrichments of gene features to predict genes underlying complex traits and diseases. medRxiv 2020:2020.2009.2008.20190561. 29. Stanzick KJ, Li Y, Schlosser P, Gorski M, Wuttke M, Thomas LF, Rasheed H, Rowan BX, Graham SE, Vanderweff BR, et al. Discovery and prioritization of variants and genes for kidney function in >1.2 million individuals. Nature Communications. 2021;12:4350. 30. The Emerging Risk Factors Collaboration, Di Angelantonio E, Sarwar N, Perry P, Kaptoge S, Ray KK, Thompson A, Wood AM, Lewington S, Sattar N, et al. Major lipids, apolipoproteins, and risk of vascular disease. JAMA. 2009;302:1993–2000. 31. Richardson TG, Sanderson E, Palmer TM, Ala-Korpela M, Ference BA, Davey Smith G, Holmes MV. Evaluating the relationship between circulating lipoprotein lipids and apolipoproteins with risk of coronary heart disease: a multivariable Mendelian randomisation analysis. PLoS Med. 2020;17:e1003062. 32. Allara E, Morani G, Carter P, Gkatzionis A, Zuber V, Foley CN, Rees JMB, Mason AM, Bell S, Gill D, et al: Genetic deter‑ minants of lipids and cardiovascular disease outcomes. Circulation: Genomic Precision Med. 2019;12:e002711. Kanoni et al. Genome Biology (2022) 23:268 Page 40 of 42 33. Veturi Y, Lucas A, Bradford Y, Hui D, Dudek S, Theusch E, Verma A, Miller JE, Kullo I, Hakonarson H, et al. A unified framework identifies new links between plasma lipids and diseases from electronic medical records across large- scale cohorts. Nat Genet. 2021;53:972–81. 34. Bush WS, Oetjens MT, Crawford DC. Unravelling the human genome-phenome relationship using phenome-wide association studies. Nat Rev Genet. 2016;17:129–45. 35. Abbott RD, Garrison RJ, Wilson PW, Epstein FH, Castelli WP, Feinleib M, LaRue C. Joint distribution of lipoprotein cholesterol classes. The Framingham study. Arteriosclerosis. 1983;3:260–72. 36. Flynn E, Tanigawa Y, Rodriguez F, Altman RB, Sinnott-Armstrong N, Rivas MA. Sex-specific genetic effects across biomarkers. Eur J Hum Genet. 2021;29:154–63. 37. Zore T, Palafox M, Reue K. Sex differences in obesity, lipid metabolism, and inflammation-A role for the sex chro‑ mosomes? Mol Metab. 2018;15:35–44. 38. AlSiraj Y, Chen X, Thatcher SE, Temel RE, Cai L, Blalock E, Katz W, Ali HM, Petriello M, Deng P, et al. XX sex chromo‑ some complement promotes atherosclerosis in mice. Nat Commun. 2019;10:2631. 39. Natarajan P, Pampana A, Graham SE, Ruotsalainen SE, Perry JA, de Vries PS, Broome JG, Pirruccello JP, Honigberg MC, Aragam K, et al. Chromosome Xq23 is associated with lower atherogenic lipid concentrations and favorable cardiometabolic indices. Nat Commun. 2021;12:2182. 40. Jiang L, Xue C, Dai S, Chen S, Chen P, Sham PC, Wang H, Li M. DESE: estimating driver tissues by selective expres‑ sion of genes associated with complex diseases or traits. Genome Biol. 2019;20:233. 41. The GTEx Consortium. The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science. 2020;369:1318–30. 42. Allara E, Morani G, Carter P, Gkatzionis A, Zuber V, Foley CN, Rees JMB, Mason AM, Bell S, Gill D, et al. Genetic deter‑ minants of lipids and cardiovascular disease outcomes: a wide-angled Mendelian randomization investigation. Circ Genom Precis Med. 2019;12:e002711. 43. Saiz-Vazquez O, Puente-Martinez A, Ubillos-Landa S, Pacheco-Bonrostro J, Santabarbara J. Cholesterol and Alzhei‑ mer’s disease risk: a meta-meta-analysis. Brain Sci. 2020;10:386. 44. Zhang X, Tian Q, Liu D, Geng T, Xu X, Ge S, Zheng D, Wu L, Song M, Hou H, et al. Causal association of circulating cholesterol levels with dementia: a mendelian randomization meta-analysis. Transl Psychiatry. 2020;10:145. 45. Tan JS, Hu MJ, Yang YM, Yang YJ. Genetic predisposition to low-density lipoprotein cholesterol may increase risks of both individual and familial Alzheimer’s disease. Front Med (Lausanne). 2021;8:798334. 46. Deb S, Puthanveetil P, Sakharkar P. A population-based cross-sectional study of the association between liver enzymes and lipid levels. Int J Hepatol. 2018;2018:1286170. 47. Joshi AD, Andersson C, Buch S, Stender S, Noordam R, Weng LC, Weeke PE, Auer PL, Boehm B, Chen C, et al. Four susceptibility loci for gallstone disease identified in a meta-analysis of genome-wide association studies. Gastro‑ enterology. 2016;151(351–363):e328. 48. Bernabeu E, Canela-Xandri O, Rawlik K, Talenti A, Prendergast J, Tenesa A. Sex differences in genetic architecture in the UK Biobank. Nat Genet. 2021;53:1283–9. 49. Ruth KS, Day FR, Tyrrell J, Thompson DJ, Wood AR, Mahajan A, Beaumont RN, Wittemans L, Martin S, Busch AS, et al. Using human genetics to understand the disease impacts of testosterone in men and women. Nat Med. 2020;26:252–8. 50. Pirastu N, Cordioli M, Nandakumar P, Mignogna G, Abdellaoui A, Hollis B, Kanai M, Rajagopal VM, Parolo PDB, Baya N, et al. Genetic analyses identify widespread sex-differential participation bias. Nat Genet. 2021;53:663–71. 51. Yengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, Frayling TM, Hirschhorn J, Yang J, Visscher PM, Consortium G. Meta-analysis of genome-wide association studies for height and body mass index in approxi‑ mately 700000 individuals of European ancestry. Hum Mol Genet. 2018;27:3641–9. 52. Bhatt DK, Basit A, Zhang H, Gaedigk A, Lee SB, Claw KG, Mehrotra A, Chaudhry AS, Pearce RE, Gaedigk R, et al. Hepatic abundance and activity of androgen- and drug-metabolizing enzyme UGT2B17 are associated with genotype, age, and sex. Drug Metab Dispos. 2018;46:888–96. 53. Nielsen JB, Rom O, Surakka I, Graham SE, Zhou W, Roychowdhury T, Fritsche LG, Gagliano Taliun SA, Sidore C, Liu Y, et al. Loss-of-function genomic variants highlight potential therapeutic targets for cardiovascular disease. Nat Commun. 2020;11:6417. 54. Aragam KG, Jiang T, Goel A, Kanoni S, Wolford BN, Weeks EM, Wang M, Hindy G, Zhou W, Grace C, et al: Discovery and systematic characterization of risk variants and genes for coronary artery disease in over a million participants. medRxiv 2021:2021.2005.2024.21257377. 55. Votava JA, Parks BW. Cross-species data integration to prioritize causal genes in lipid metabolism. Curr Opin Lipidol. 2021;32:141–6. 56. Kherallah RY, Khawaja M, Olson M, Angiolillo D, Birnbaum Y. Cilostazol: a review of basic mechanisms and clinical uses. Cardiovasc Drugs Ther. 2022;36:777-92. 57. Sabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, Kuder JF, Wang H, Liu T, Wasserman SM, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N Engl J Med. 2017;376:1713–22. 58. Schwartz GG, Steg PG, Szarek M, Bhatt DL, Bittner VA, Diaz R, Edelberg JM, Goodman SG, Hanotin C, Harrington RA, et al. Alirocumab and cardiovascular outcomes after acute coronary syndrome. N Engl J Med. 2018;379:2097–107. 59. Ray KK, Wright RS, Kallend D, Koenig W, Leiter LA, Raal FJ, Bisch JA, Richardson T, Jaros M, Wijngaard PLJ, Kastelein JJP. Two phase 3 trials of inclisiran in patients with elevated LDL cholesterol. N Engl J Med. 2020;382:1507–19. 60. Nelson CP, Lai FY, Nath M, Ye S, Webb TR, Schunkert H, Samani NJ. Genetic assessment of potential long-term on-target side effects of PCSK9 (Proprotein Convertase Subtilisin/Kexin Type 9) inhibitors. Circ Genom Precis Med. 2019;12:e002196. 61. Mahajan A, Taliun D, Thurner M, Robertson NR, Torres JM, Rayner NW, Payne AJ, Steinthorsdottir V, Scott RA, Grarup N, et al. Fine-mapping type 2 diabetes loci to single-variant resolution using high-density imputation and islet- specific epigenome maps. Nat Genet. 2018;50:1505–13. 62. Assimes T, Catherine T, Xiang Z, Austin H, Shoa C, Valerio N, Shining M, Huaying F, Bryan RG, Kyung Min L, et al. A large-scale multi-ethnic genome-wide association study of coronary artery disease. Nat Med. 2022;28:1679-92. Kanoni et al. Genome Biology (2022) 23:268 Page 41 of 42 63. Ridker PM, Revkin J, Amarenco P, Brunell R, Curto M, Civeira F, Flather M, Glynn RJ, Gregoire J, Jukema JW, et al. Cardiovascular efficacy and safety of bococizumab in high-risk patients. N Engl J Med. 2017;376:1527–39. 64. Hopewell JC, Malik R, Valdes-Marquez E, Worrall BB, Collins R. ISGC MCot: Differential effects of PCSK9 variants on risk of coronary disease and ischaemic stroke. Eur Heart J. 2018;39:354–9. 65. Hayat S, Hassan Z, Changazi SH, Zahra A, Noman M. Zain Ul Abdin M, Javed H, Ans AH: Comparative analysis of serum lipid profiles in patients with and without gallstones: a prospective cross-sectional study. Ann Med Surg (Lond). 2019;42:11–3. 66. Wang J, Shen S, Wang B, Ni X, Liu H, Ni X, Yu R, Suo T, Liu H. Serum lipid levels are the risk factors of gallbladder stones: a population-based study in China. Lipids Health Dis. 2020;19:50. 67. Gustafsson U, Sahlin S, Einarsson C. Biliary lipid composition in patients with cholesterol and pigment gallstones and gallstone-free subjects: deoxycholic acid does not contribute to formation of cholesterol gallstones. Eur J Clin Invest. 2000;30:1099–106. 68. Weerakoon HT, Ranasinghe S, Navaratne A, Sivakanesan R, Galketiya KB, Rosairo S. Serum lipid concentrations in patients with cholesterol and pigment gallstones. BMC Res Notes. 2014;7:548. 69. Chen L, Yang H, Li H, He C, Yang L, Lv G. Insights into modifiable risk factors of cholelithiasis: a Mendelian randomi‑ zation study. Hepatology. 2022;75:785–96. 70. Yu XH, Qian K, Jiang N, Zheng XL, Cayabyab FS, Tang CK. ABCG5/ABCG8 in cholesterol excretion and atherosclero‑ sis. Clin Chim Acta. 2014;428:82–8. 71. Silbernagel G, Chapman MJ, Genser B, Kleber ME, Fauler G, Scharnagl H, Grammer TB, Boehm BO, Makela KM, Kahonen M, et al. High intestinal cholesterol absorption is associated with cardiovascular disease and risk alleles in ABCG8 and ABO: evidence from the LURIC and YFS cohorts and from a meta-analysis. J Am Coll Cardiol. 2013;62:291–9. 72. Teupser D, Baber R, Ceglarek U, Scholz M, Illig T, Gieger C, Holdt LM, Leichtle A, Greiser KH, Huster D, et al. Genetic regulation of serum phytosterol levels and risk of coronary artery disease. Circulation Cardiovasc Genet. 2010;3:331–9. 73. Artha I, Bhargah A, Dharmawan NK, Pande UW, Triyana KA, Mahariski PA, Yuwono J, Bhargah V, Prabawa IPY, Manu‑ aba I, Rina IK. High level of individual lipid profile and lipid ratio as a predictive marker of poor glycemic control in type-2 diabetes mellitus. Vasc Health Risk Manag. 2019;15:149–57. 74. Hussain A, Ali I, Ijaz M, Rahim A. Correlation between hemoglobin A1c and serum lipid profile in Afghani patients with type 2 diabetes: hemoglobin A1c prognosticates dyslipidemia. Ther Adv Endocrinol Metab. 2017;8:51–7. 75. Chen J, Spracklen CN, Marenne G, Varshney A, Corbin LJ, Luan J, Willems SM, Wu Y, Zhang X, Horikoshi M, et al. The trans-ancestral genomic architecture of glycemic traits. Nat Genet. 2021;53:840–60. 76. Chami N, Chen MH, Slater AJ, Eicher JD, Evangelou E, Tajuddin SM, Love-Gregory L, Kacprowski T, Schick UM, Nomura A, et al. Exome genotyping identifies pleiotropic variants associated with red blood cell traits. Am J Hum Genet. 2016;99:8–21. 77. Leong A, Chen J, Wheeler E, Hivert MF, Liu CT, Merino J, Dupuis J, Tai ES, Rotter JI, Florez JC, et al. Mendelian rand‑ omization analysis of hemoglobin A1c as a risk factor for coronary artery disease. Diabetes Care. 2019;42:1202–8. 78. McDonagh EM, Thorn CF, Bautista JM, Youngster I, Altman RB, Klein TE. PharmGKB summary: very important phar‑ macogene information for G6PD. Pharmacogenet Genomics. 2012;22:219–28. 79. Dore MP, Parodi G, Portoghese M, Pes GM. The controversial role of glucose-6-phosphate dehydrogenase defi‑ ciency on cardiovascular disease: a narrative review. Oxid Med Cell Longev. 2021;2021:5529256. 80. Spielman RS, Bastone LA, Burdick JT, Morley M, Ewens WJ, Cheung VG. Common genetic variants account for dif‑ ferences in gene expression among ethnic groups. Nat Genet. 2007;39:226–31. 81. Zhu AZ, Cox LS, Ahluwalia JS, Renner CC, Hatsukami DK, Benowitz NL, Tyndale RF. Genetic and phenotypic varia‑ tion in UGT2B17, a testosterone-metabolizing enzyme, is associated with BMI in males. Pharmacogenet Genom‑ ics. 2015;25:263–9. 82. Yang TL, Chen XD, Guo Y, Lei SF, Wang JT, Zhou Q, Pan F, Chen Y, Zhang ZX, Dong SS, et al. Genome-wide copy-num‑ ber-variation study identified a susceptibility gene, UGT2B17, for osteoporosis. Am J Hum Genet. 2008;83:663–74. 83. Gencer B, Bonomi M, Adorni MP, Sirtori CR, Mach F, Ruscica M. Cardiovascular risk and testosterone - from subclini‑ cal atherosclerosis to lipoprotein function to heart failure. Rev Endocr Metab Disord. 2021;22:257–74. 84. Firtser S, Juonala M, Magnussen CG, Jula A, Loo BM, Marniemi J, Viikari JS, Toppari J, Perheentupa A, Hutri-Kahonen N, Raitakari OT. Relation of total and free testosterone and sex hormone-binding globulin with cardiovascular risk factors in men aged 24–45 years. The Cardiovascular Risk in Young Finns Study. Atherosclerosis. 2012;222:257–62. 85. Schooling CM, Luo S, Au Yeung SL, Thompson DJ, Karthikeyan S, Bolton TR, Mason AM, Ingelsson E, Burgess S. Genetic predictors of testosterone and their associations with cardiovascular disease and risk factors: a Mendelian randomization investigation. Int J Cardiol. 2018;267:171–6. 86. Au Yeung SL, Cheng KK, Zhao J, Zhang W, Jiang C, Lam TH, Leung GM, Schooling CM. Genetically predicted 17beta-estradiol and cardiovascular risk factors in women: a Mendelian randomization analysis using young women in Hong Kong and older women in the Guangzhou Biobank Cohort Study. Ann Epidemiol. 2016;26:171–5. 87. Winkler TW, Day FR, Croteau-Chonka DC, Wood AR, Locke AE, Magi R, Ferreira T, Fall T, Graff M, Justice AE, et al. Quality control and conduct of genome-wide association meta-analyses. Nat Protoc. 2014;9:1192–212. 88. Magi R, Horikoshi M, Sofer T, Mahajan A, Kitajima H, Franceschini N, McCarthy MI, Cogent-Kidney Consortium TDGC, Morris AP. Trans-ethnic meta-regression of genome-wide association studies accounting for ancestry increases power for discovery and improves fine-mapping resolution. Hum Mol Genet. 2017;26:3639–50. 89. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformat‑ ics. 2010;26:2190–1. 90. Liu DJ, Peloso GM, Zhan X, Holmen OL, Zawistowski M, Feng S, Nikpay M, Auer PL, Goel A, Zhang H, et al. Meta- analysis of gene-level tests for rare variant association. Nat Genet. 2014;46:200–4. 91. Winkler TW, Justice AE, Cupples LA, Kronenberg F, Kutalik Z, Heid IM. consortium G: Approaches to detect genetic effects that differ between two strata in genome-wide meta-analyses: recommendations based on a systematic evaluation. PLoS ONE. 2017;12:e0181038. 92. Fauman EB, Hyde C: An optimal variant to gene distance window derived from an empirical definition of cis and trans protein QTLs. bioRxiv 2022:2022.2003.2007.483314. Kanoni et al. Genome Biology (2022) 23:268 Page 42 of 42 93. Giambartolomei C, Vukcevic D, Schadt EE, Franke L, Hingorani AD, Wallace C, Plagnol V. Bayesian test for colocalisa‑ tion between pairs of genetic association studies using summary statistics. PLoS Genet. 2014;10:e1004383. 94. Caliskan M, Manduchi E, Rao HS, Segert JA, Beltrame MH, Trizzino M, Park Y, Baker SW, Chesi A, Johnson ME, et al. Genetic and epigenetic fine mapping of complex trait associated loci in the human liver. Am J Hum Genet. 2019;105:89–107. 95. Barbeira AN, Dickinson SP, Bonazzola R, Zheng J, Wheeler HE, Torres JM, Torstenson ES, Shah KP, Garcia T, Edwards TL, et al. Exploring the phenotypic consequences of tissue specific gene expression variation inferred from GWAS summary statistics. Nat Commun. 1825;2018:9. 96. Hindy G, Dornbos P, Chaffin MD, Liu DJ, Wang M, Selvaraj MS, Zhang D, Park J, Aguilar-Salinas CA, Antonacci-Fulton L, et al. Rare coding variants in 35 genes associate with circulating lipid levels-a multi-ancestry analysis of 170,000 exomes. Am J Hum Genet. 2022;109:81–96. 97. Brown EE, Sturm AC, Cuchel M, Braun LT, Duell PB, Underberg JA, Jacobson TA, Hegele RA. Genetic testing in dyslipidemia: a scientific statement from the National Lipid Association. J Clin Lipidol. 2020;14:398–413. 98. Hegele RA, Boren J, Ginsberg HN, Arca M, Averna M, Binder CJ, Calabresi L, Chapman MJ, Cuchel M, von Eckard‑ stein A, et al. Rare dyslipidaemias, from phenotype to genotype to management: a European Atherosclerosis Society task force consensus statement. Lancet Diabetes Endocrinol. 2020;8:50–67. 99. Zhou Y, Zhang Y, Lian X, Li F, Wang C, Zhu F, Qiu Y, Chen Y. Therapeutic target database update 2022: facilitating drug discovery with enriched comparative data of targeted agents. Nucleic Acids Res. 2022;50:D1398–407. 100. Ge T, Chen CY, Ni Y, Feng YA, Smoller JW. Polygenic prediction via Bayesian regression and continuous shrinkage priors. Nat Commun. 2019;10:1776. 101. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, Sham PC. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81:559–75. 102. Denny JC, Ritchie MD, Basford MA, Pulley JM, Bastarache L, Brown-Gentry K, Wang D, Masys DR, Roden DM, Craw‑ ford DC. PheWAS: demonstrating the feasibility of a phenome-wide scan to discover gene-disease associations. Bioinformatics. 2010;26:1205–10. 103. Verma A, Bradford Y, Dudek S, Lucas AM, Verma SS, Pendergrass SA, Ritchie MD. A simulation study investigating power estimates in phenome-wide association studies. BMC Bioinformatics. 2018;19:120. 104. Nelson CP, Goel A, Butterworth AS, Kanoni S, Webb TR, Marouli E, Zeng L, Ntalla I, Lai FY, Hopewell JC, et al. Association analyses based on false discovery rate implicate new loci for coronary artery disease. Nat Genet. 2017;49:1385–91. 105. Liu Z, Zhang Y, Graham S, Wang X, Cai D, Huang M, Pique-Regi R, Dong XC, Chen YE, Willer C, Liu W. Causal relation‑ ships between NAFLD, T2D and obesity have implications for disease subphenotyping. J Hepatol. 2020;73:263–76. 106. Zhou W, Nielsen JB, Fritsche LG, Dey R, Gabrielsen ME, Wolford BN, LeFaive J, VandeHaar P, Gagliano SA, Gifford A, et al. Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies. Nat Genet. 2018;50:1335–41. 107. Graham SE, Clarke SL, Wu KH, Kanoni S, Zajac GJM, Ramdas S, Surakka I, Ntalla I, Vedantam S, Winkler TW, et al: GLGC GWAS meta-analysis results and risk score weights repository: http://​csg.​sph.​umich.​edu/​willer/​public/​glgc-​lipid​s2021 2021. 108. Kanoni S GS, Wang Y, Surakka I, Ramdas S, Zhu X, Costanzo M, Jang D, Burtt NP, Willer CJ, Assimes TL, Peloso GM: A web browser displaying the gene prioritization and PheWAS results: https://​hugea​mp.​org:​8000/​resea​rch.​html?​ pageid=​GLGC_​149 2021. 109. Graham SE, Clarke SL, Wu KH, Kanoni S, Zajac GJM, Ramdas S, Surakka I, Ntalla I, Vedantam S, Winkler TW, et al: Optimized trans-ancestry polygenic score weights for LDL in the PGS Catalog : https://​www.​pgsca​talog.​org/​publi​ cation/​PGP00​0230/ 2021. 110. Kanoni S, Graham SE, Wang Y, Surakka I, Ramdas S, Zhu X, Clarke SL, Bhatti KF, Vedantam S, Winkler TW, et al: Optimized trans-ancestry polygenic score weights for HDL, TC, TG and non-HDL in the PGS Catalog: https://​www.​ pgsca​talog.​org/​publi​cation/​PGP00​0366/ 2022. 111. Kanoni S, Graham SE, Wang Y, Surakka I, Ramdas S, Zhu X, Clarke SL, Bhatti KF, Vedantam S, Winkler TW, et al: Impli‑ cating genes, pleiotropy and sexual dimorphism at blood lipid loci through multi-ancestry meta-analysis. Github. https://​github.​com/​Global-​Lipids-​Genet​ics. 2022. 112. Kanoni S, Graham SE, Wang Y, Surakka I, Ramdas S, Zhu X, Clarke SL, Bhatti KF, Vedantam S, Winkler TW, et al. Impli‑ cating genes, pleiotropy and sexual dimorphism at blood lipid loci through multi-ancestry meta-analysis. 2022. Zenodo. https://​doi.​org/​10.​5281/​zenodo.​71302​99. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.