BRIEF RESEARCH REPORT published: 08 July 2022 doi: 10.3389/fcvm.2022.940696 A Polygenic Risk Score Based on a Edited by: Zhihua Wang, Cardioembolic Stroke Multitrait Chinese Academy of Medical Sciences and Peking Union Medical Analysis Improves a Clinical College, China Reviewed by: Prediction Model for This Stroke Xiao Chang, Children’s Hospital of Philadelphia, Subtype United States Qingqing Yan, Jara Cárcel-Márquez 1,2† , Elena Muiño 1† , Cristina Gallego-Fabrega 1,3 , Natalia Cullell 1,4 , Chinese Academy of Medical Miquel Lledós 1 , Laia Llucià-Carol 1,5 , Tomás Sobrino 6 , Francisco Campos 6 , José Castillo 6 , Sciences and Peking Union Medical Marimar Freijo 7 , Juan Francisco Arenillas 8 , Victor Obach 9 , José Álvarez-Sabín 10 , College, China Georgios Tsivgoulis, Carlos A. Molina 10 , Marc Ribó 10 , Jordi Jiménez-Conde 11 , Jaume Roquer 11 , National and Kapodistrian University Lucia Muñoz-Narbona 12 , Elena Lopez-Cancio 13 , Mònica Millán 12 , Rosa Diaz-Navarro 14 , of Athens, Greece Cristòfol Vives-Bauza 14 , Gemma Serrano-Heras 15 , Tomás Segura 15 , Laura Ibañez 16 , Laura Heitsch 17,18 , Pilar Delgado 19 , Rajat Dhar 18 , Jerzy Krupinski 20 , *Correspondence: Israel Férnandez-Cadenas Raquel Delgado-Mederos 3 , Luis Prats-Sánchez 3 , Pol Camps-Renom 3 , Natalia Blay 21 , israelcadenas@yahoo.es Lauro Sumoy 22 , Rafael de Cid 21 , Joan Montaner 23 , Carlos Cruchaga 16,24 , Jin-Moo Lee 18 , Joan Martí-Fàbregas 3 and Israel Férnandez-Cadenas 1* † These authors have contributed on behalf of GeneStroke Consortium and International Stroke Genetics Consortium equally to this work 1 Stroke Pharmacogenomics and Genetics Group, Institut d’Investigació Biomèdica Sant Pau (IIB SANT PAU), Barcelona, Specialty section: Spain, 2 Departament de Medicina, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Stroke Unit, Department of This article was submitted to Neurology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 4 Stroke Pharmacogenomics and Genetics Laboratory, Cardiovascular Genetics and Systems Fundación Docència i Recerca Mútua Terrassa, Hospital Mútua Terrassa, Terrassa, Spain, 5 Department de Genética i de Medicine, Microbiologia, Universitat Autónoma de Barcelona, Barcelona, Spain, 6 Clinical Neurosciences Research Laboratory, Hospital a section of the journal Clínico Universitario de Santiago de Compostela, Health Research Institute of Santiago de Compostela (IDIS), Santiago de Frontiers in Cardiovascular Medicine Compostela, Spain, 7 Department of Neurology, Biocruces-Bizkaia Health Research Institute, Bilbao, Spain, 8 Stroke Unit, Department of Neurology, University Hospital of Valladolid, Valladolid, Spain, 9 Department of Neurology, Hospital Clínic de Received: 10 May 2022 Barcelona, IDIBAPS, Barcelona, Spain, 10 Stroke Unit, Department of Neurology, Hospital Universitari Vall d’Hebron, Accepted: 06 June 2022 Barcelona, Spain, 11 Department of Neurology, IMIM-Hospital del Mar; Neurovascular Research Group, IMIM (Institut Hospital Published: 08 July 2022 del Mar d’Investigacions Mèdiques), Universitat Autònoma de Barcelona/DCEXS-Universitat Pompeu Fabra, Barcelona, Citation: Spain, 12 Department of Neurosciences, Hospital Germans Trias i Pujol, Universitat Autònoma de Barcelona, Barcelona, Cárcel-Márquez J, Muiño E, Spain, 13 Department of Neurology, University Hospital Central de Asturias (HUCA). Oviedo, Spain, 14 Department of Gallego-Fabrega C, Cullell N, Neurology, Son Espases University Hospital, Illes Balears Health Research Institute (IdISBa), Palma, Spain, 15 Department of Lledós M, Llucià-Carol L, Sobrino T, Neurology, University Hospital of Albacete, Albacete, Spain, 16 Department of Psychiatry, Washington University School of Campos F, Castillo J, Freijo M, Medicine, Saint Louis, MO, United States, 17 Department of Emergency Medicine, Washington University School of Medicine, Arenillas JF, Obach V, Álvarez-Sabín J, Saint Louis, MO, United States, 18 Department of Neurology, Washington University School of Medicine, Saint Louis, MO, Molina CA, Ribó M, Jiménez-Conde J, United States, 19 Neurovascular Research Laboratory, Vall d’Hebron Institute of Research, Universitat Autònoma de Roquer J, Muñoz-Narbona L, Barcelona, Barcelona, Spain, 20 Neurology Service, Hospital Universitari Mútua Terrassa, Terrassa, Spain, Lopez-Cancio E, Millán M, 21 GenomesForLife-GCAT Lab, Germans Trias i Pujol Research Institute (IGTP), Badalona, Spain, 22 High Content Genomics Diaz-Navarro R, Vives-Bauza C, and Bioinformatics Unit, Germans Trias i Pujol Research Institute (IGTP), Badalona, Spain, 23 Institute de Biomedicine of Serrano-Heras G, Segura T, Ibañez L, Seville, IBiS/Hospital Universitario Virgen del Rocío/CSIC/University of Seville and Department of Neurology, Hospital Heitsch L, Delgado P, Dhar R, Universitario Virgen Macarena, Seville, Spain, 24 Neurogenomics and Informatics Center at Washington University in St. Louis, Krupinski J, Delgado-Mederos R, Saint Louis, MO, United States Prats-Sánchez L, Camps-Renom P, Blay N, Sumoy L, de Cid R, Montaner J, Cruchaga C, Lee J-M, Background: Occult atrial fibrillation (AF) is one of the major causes of embolic stroke Martí-Fàbregas J and of undetermined source (ESUS). Knowing the underlying etiology of an ESUS will Fernandez-Cadenas I (2022) A reduce stroke recurrence and/or unnecessary use of anticoagulants. Understanding Polygenic Risk Score Based on a Cardioembolic Stroke Multitrait cardioembolic strokes (CES), whose main cause is AF, will provide tools to select patients Analysis Improves a Clinical Prediction who would benefit from anticoagulants among those with ESUS or AF. We aimed to Model for This Stroke Subtype. Front. Cardiovasc. Med. 9:940696. discover novel loci associated with CES and create a polygenetic risk score (PRS) for a doi: 10.3389/fcvm.2022.940696 more efficient CES risk stratification. Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction Methods: Multitrait analysis of GWAS (MTAG) was performed with MEGASTROKE- CES cohort (n = 362,661) and AF cohort (n = 1,030,836). We considered significant variants and replicated those variants with MTAG p-value < 5 × 10−8 influencing both traits (GWAS-pairwise) with a p-value < 0.05 in the original GWAS and in an independent cohort (n = 9,105). The PRS was created with PRSice-2 and evaluated in the independent cohort. Results: We found and replicated eleven loci associated with CES. Eight were novel loci. Seven of them had been previously associated with AF, namely, CAV1, ESR2, GORAB, IGF1R, NEURL1, WIPF1, and ZEB2. KIAA1755 locus had never been associated with CES/AF, leading its index variant to a missense change (R1045W). The PRS generated has been significantly associated with CES improving discrimination and patient reclassification of a model with age, sex, and hypertension. Conclusion: The loci found significantly associated with CES in the MTAG, together with the creation of a PRS that improves the predictive clinical models of CES, might help guide future clinical trials of anticoagulant therapy in patients with ESUS or AF. Keywords: polygenic risk score, GWAS, multi-trait analysis, stroke, ESUs INTRODUCTION Association Study (MTAG) of CES-AF and create a polygenetic risk score (PRS) that allowed a more efficient stratification of About 25% of ischemic strokes are of undetermined etiology (1): stroke patient risk of having a CES. patients with multiple stroke etiologies, incomplete diagnostic work-up, or embolic stroke of undetermined source (ESUS). Up to 17% of all ischemic strokes are ESUS, with a stroke recurrence METHODS rate of 4–5% despite antiplatelet therapy (2). The data that support the findings of this study are available from The ESUS encompasses different entities. Atrial cardiopathy, the corresponding author upon reasonable request. occult atrial fibrillation (AF), and left ventricular disease might benefit from anticoagulation, but atherosclerotic plaques might benefit from low-dose anticoagulation with antiplatelets in ESUS Cohorts’ Description The summary statistics for CES were obtained from the patients (2). The subgroup of patients >75 years in RE-SPECT MEGASTROKE analysis (MEGASTROKE-CES) through ESUS (Dabigatran Etexilate for Secondary Stroke Prevention the Cerebrovascular Disease Knowledge Portal (http:// in Patients With Embolic Stroke of Undetermined Source) cerebrovascularportal.org). This cohort was composed of had a significant benefit of lower-dose dabigatran over aspirin, 7,193 CES patients and 355,468 controls of European ancestry. suggesting occult AF as a triggering cause (3). Different studies The summary statistics for AF were obtained from the Atrial indicate that the prevalence of occult AF among ESUS patients is Fibrillation 2018 (AF-2018) analysis through the GWAS catalog 11–30% (2, 4). portal (https://www.ebi.ac.uk/gwas/). The AF-2018 cohort A tool capable of better stratifying patients is needed to offer was composed of 60,620 AF cases and 970,216 controls. The them appropriate treatment regarding its potential stroke cause characteristics of the individuals in both studies are listed in the to decrease its recurrence. Supplementary Material and their respective publications (6, 7). On the contrary, not all patients with AF will develop a stroke, and the decision of anticoagulation for stroke prevention in AF patients is carried out based on a clinical scale: CHA2 DS2 -VASc. Single-Nucleotide Variation Quality The rates of stroke vary considerably in patients with CHA2 DS2 - Controls VASc 1–2 (5) and hence the need for a more accurate scale in A series of standard quality controls (QC) was applied to select these cases. the single-nucleotide variants (SNVs) for the analysis. Variant Cardioembolic strokes (CES) are mostly caused by exclusion criteria include the following (1): Not common to an onset/already known AF. Understanding CES genetic the summary statistics of the traits (2), Minor allele frequency architecture will provide tools to select ESUS or AF patients lower or equal to 0.01 (3), Missing values (4), Negative standard who would benefit from anticoagulants and develop specific and error or not a number value (5), p-value of 0, 6 Not SNVs (7), more effective therapies with fewer side effects. Duplicated SNVs (8), Strand ambiguity, and (8) Inconsistent Therefore, we aimed to discover novel loci associated with allele pairs. Locus 15q21.3, which prioritized genes GCOM1 and CES by performing a Multitrait Analysis of Genome Wide MYZAP from AF-2018, was not evaluated due to the absence Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction of the significant SNVs of AF-2018 in the MEGASTROKE- (Supplementary Table 2). Standard QCs were performed using CES analysis. the PLINK v1.9 and KING v2.1.3 software. Imputation was performed in the Michigan Imputation Server Pipeline (12) using Multitrait Analysis of GWAS Minimac4 and HRC r1.1 2016 panel. Further descriptions of QCs We applied MTAG (8) of MEGASTROKE-CES and AF-2018 and imputation are present in the Supplementary Material. summary statistics. We considered loci to be significantly associated with the trait of interest when the p-value was < 5 GWAS Analysis × 10−8 in the MTAG result and the p-value was < 0.05 in We performed two different GWAS in the same cohort (for the the original GWAS. We considered replicating the SNVs with a two different traits here studied), with an additive genetic model p-value < 0.05 in the GWAS of our independent cohort. using fastGWA from GCTA (13). We studied the association To avoid an increase in the type I error rate due to the presence with CES (CES = 1,515; controls = 5,626) and AF (AF patients of SNVs that are not associated with CES but with AF or vice = 1,110; controls = 7,791). Age, sex, and the first 10 principal versa, we used GWAS-pairwise (9). This is a Bayesian pleiotropy components were used as covariates. association test to identify genetic variants that influence pairs The results of these two GWAS were used to evaluate of traits (9). We used it to ensure that the leading SNV of a replicability. We studied those index variants from significant loci significant locus belongs to a genomic region influenced by both with a p-value < 5 × 10−8 in the MTAG, a p-value < 0.05 in the traits evaluated (9), since SNPs that are not really associated with original GWAS used for performing the MTAG, and PPA-3 > 0.6 one trait, but are associated with the other one, could bias effect- that suggests that the genomic region is associated with CES and size estimates for the first trait and increase false-positive rate (8). AF. We considered the replicated SNVs with a p-value < 0.05 and The posterior probability for model-3 (PPA-3) >0.6 suggests that a consistent direction of the effect on this analysis. a specific genomic region is associated with both traits. A PPA- 1 >0.6 will suggest that the genomic region is associated only Functional Annotation and Gene with CES, and a PPA-2 >0.6 is associated only with AF. Genomic inflation was estimated as lambda. Prioritization Gene prioritization was performed for the novel loci using Identification of Independent and Novel Variant-to-Gene tool from Open targets Genetics Version 7 (14). This tool integrates biological evidence of four main data Loci Associated With CES types, namely, (1) molecular phenotype quantitative trait loci Independent loci were defined as those >1 megabase (Mb) experiments (QTLs), (2) chromatin interaction experiments, apart in the physical distance among SNVs with a genome-wide e.g., Promoter Capture Hi-C (PCHi-C), 3) in silico functional significance threshold of p-value < 5 × 10−86 . Loci were defined predictions, e.g., Variant Effect Predictor (VEP) from Ensembl as novel when SNVs had an r2 < 0.1 compared with the index and (4) distance between the variant and each gene’s canonical SNVs of the loci, namely, PITX2 7 , ZFHX3 7 , NKX2-5 7 , RGS7 7 , transcription start site (TSS). Additionally, we used the HaploReg ABO 7 10 , PHF20 11 , GNAO1 11 , and 5q22.3 11 , that were GWAS database to determine the functional annotation of the most significant in previous studies. strongly associated SNVs per locus. For the missense SNVs, we determined the likelihood that amino acid substitution has a Replication Stage in an Independent deleterious effect on protein function using SIFT. European Cohort We performed GWAS in an independent cohort of 9,105 individuals [GENERACION cohort: 3,479 ischemic stroke (IS) Gene Set Analysis We conducted a WebGestalt Overrepresentation Analysis of patients and 5,625 controls]. IS patients over 18 years were the selected prioritized genes associated with MTAG-CES. Gene recruited via hospital-based studies, between 2003 and 2020 Ontology (GO) of biological processes was performed, as well in Spain, if they had a measurable neurologic deficit on the as a Benjamini Hochberg correction of the association p-value. NIHSS within 6 h of the last known asymptomatic status and had We defined a biological process with a p-value < 0.05 as been diagnosed with stroke by an experienced neurologist and statistically significant. confirmed by neuroimaging (10, 11). Controls were subjects over 18 years recruited in Spain, without a history of IS, who declared they were free of neurovascular diseases before enrollment. An Polygenic Risk Score Development Institutional Review Board or Ethics Committee approved the A PRS was conducted through the PRSice-2 software version study at each participating site. All patients or their relatives 2.3.3 (15), where the estimation is based on the risk alleles of provided written informed consent. Further description of the having a CES and their effect size extracted from the regions cohorts is present in Supplementary Material, as well as the with PPA-3 > 0.6 of the MTAG summary statistics created in array information, the contribution of hospitals, and the clinical this study. description (Supplementary Tables 1–3). GENERACION cohort was randomly split into training and test sets in 80:20 proportion. Best score threshold selection was Quality Control and Imputation performed based on the major variance explained by the score The DNA samples were genotyped on commercial arrays from (PRS r2 ) in the training set. The evaluation of this score was Illumina R (San Diego, CA) and AxiomTM Spain Biobank array performed in the independent test set. Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction FIGURE 1 | Workflow of the SNVs for the AF-2018 and MEGASTROKE-CES datasets. NaN, Not a number; SNV, Single-nucleotide variant. We used R version 4.1.3 and Bioconductor packages to Additionally, we estimated the AUC and AUPRC for each evaluate the clinical relevance of this PRS. We calculated individual predictor. three models, namely, model-1 including only the PRS; model-2 including statistically significant clinical variables with <10% missing values, since a high rate of missing RESULTS values might bias the results of subsequent statistical MTAG Analysis of CES analyses (16); and model-3 adding the PRS to the model-2. After QCs (Figure 1), there were 6,808,676 common qualified Model discrimination was assessed with the area under the SNVs from the AF-2018 and MEGASTROKE-CES cohorts. ROC curve (AUC) and the area under the precision recall The MTAG software revealed mean χ2 for AF-2018 and curve (AUPRC). We used DeLong’s test for two correlated MEGASTROKE-CES in 1.39 and 1.12, respectively. The ROC curves to find out whether there are significant estimated equivalent GWAS sample size of the MTAG analysis differences between the discrimination of the models. The for CES was 861,823 individuals. A Manhattan plot of the MTAG- net reclassification index (NRI) and integrated discrimination CES analysis is shown in Figure 2; less evidence of genomic index (IDI) were performed to evaluate model-2 and model-3. inflation was observed with a lambda of 1.02. Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction FIGURE 2 | Manhattan plot of MTAG-CES. The X-axis represents chromosome location, and the Y-axis represents the minus logarithm on base 10 of p-value. The red line represents the GWAS-significance threshold. The novel loci are shown in blue and the established loci are shown in yellow. The MTAG-CES results revealed a total of 44 associated loci to genome-wide significant loci from the MTAG-CES revealed (p-value < 5 × 10−8 ); 40 significant loci associated with CES that 11 loci were replicated, as SNVs had a p-value < 0.05 in were novel, and four were previously found known associations MEGASTROKE-CES, a PPA-3 > 0.6, and a p-value < 0.05 in the (Table 1). All loci significantly associated with MEGASTROKE- replication cohort (GENERACION). Among these 11 different CES (ABO, NKX2-5, PITX2, and ZFHX3) were genome-wide loci, PITX2, ZFHX3, and NKX2-5 were already known. Eight associated in this MTAG-CES, except for the locus belonging loci were novel associations whose prioritized genes were CAV1, to RGS7 gene (top SNV rs146390073 MTAG p-value = 0.001, IGF1R, KIAA1755, NEURL1, GORAB, ESR2, ZEB2, and WIPF1 AF-2018 p-value = 0.98). (Supplementary Table 5). Other loci found significant in previous GWAS of CES Interestingly, we found loci not previously described in AF or different from MEGASTROKE were: PHF20, GNAO1, CES with four prioritized genes, namely, TMEM60, KIAA1755, and 5q22.3 region. For two of them, the association was NCOR2, and FILIP1. Functional annotation of the index SNVs more significant in our analysis. For the 5q22.3 region, top revealed rs3746471 as a missense variant of the KIAA1755 gene SNV rs2169955 MTAG-CES p-value = 4.76 × 10−7 vs. coding for R1045W, and it was predicted to be deleterious with a MEGASTROKE-CES p-value = 6.13 × 10−3 , and for mapped SIFT score of 0.007. gene PHF20, top SNV rs11697087 MTAG-CES p-value = 6.62 × 10−5 vs. MEGASTROKE-CES p-value = 6.05 × 10−4 . GNAO1 was not evaluated in our study due to the absence of the New Candidate Locus Associated With AF index SNV in AF-2018. FILIP1 locus reached genome-wide significance in the MTAG- AF, a p-value < 0.05 in AF-2018 and a PPA-3 > 0.6. The New Candidate Loci Associated With CES FILIP1 index variant was additionally evaluated in the GWAS After gene prioritization, 44 genes were selected from of AF in the independent cohort (GENERACION), revealing a the 44 loci (Supplementary Table 4). Novel loci showed suggestive p-value with a consistent direction of the effect of this a high degree of functionality of the SNVs as missense novel association with AF rs12211255-A, beta(se) = 0.013(0.007), variants, eQTL, pQTLs, and HiC physical interaction p-value = 0.09. (Supplementary Table 4). The study of the 111 AF-2018 significant loci Replication analysis was performed in a new cohort of (Supplementary Table 6) using GWAS-pairwise strategy IS patients and controls (GENERACION cohort, n = 9,105). suggested 51 loci that have an exclusive association with AF risk Evaluation of the index SNVs and SNVs in high LD belonging and a lack of association with CES. Frontiers in Cardiovascular Medicine | www.frontiersin.org 5 July 2022 | Volume 9 | Article 940696 Frontiers in Cardiovascular Medicine | www.frontiersin.org Cárcel-Márquez et al. TABLE 1 | MTAG-CES results of the independent and significant loci. MEGASTROKE CES AF-2018 MTAG-CES MTAG-AF SNV Locus Gene Novelty Z P Z P Z P Z P PPA rs17042098-A 4q25 PITX2 (Malik et 12.72 3.72 x 10−37 38.01 8.97 x 32.86 7.73 x 37.93 0 1.00 al.) [7] 10−318 10−237 rs2106261-T 16q22.3 ZFHX3 (Malik et 6.75 1.63 x 10−11 20.23 4.97 x 17.48 2.02 x 20.19 1.20 x 1.00 al.) [7] 10−91 10−68 10−90 rs11264280-T 1q21.3 ADAM15 Novel 2.49 1.28 x 10−02 18.97 3.07 x 14.22 7.17 x 18.44 5.88 x 0.65 10−79 10−46 10−76 rs11598047-A 10q24.33 NEURL1 Novel −3.27 1.06 x 10−03 −17.08 8.95 x −13.38 7.50 x −16.73 7.42 x 0.93 10−66 10−41 10−63 rs3807989-A 7q31.2 CAV1 Novel −4.50 6.75 x 10−06 −15.63 1.24 x −13.10 3.43 x −15.50 3.30 x 1.00 10−54 10−39 10−54 rs680084-A 1q24.2 GORAB Novel −4.10 4.02 x 10−05 −13.54 3.31 x −11.46 2.10 x −13.46 2.84 x 0.99 10−42 10−30 10−41 rs883079-T 12q24.21 TBX5 Novel 3.55 3.95 x 10−04 13.26 2.84 x 10.96 5.97 x 13.12 2.60 x 0.97 10−40 10−28 10−39 rs17171711-T 5q31.2 FAM13B Novel 3.79 1.56 x 10−04 12.48 1.95 x 10.57 4.04 x 12.41 2.35 x 0.99 10−35 10−26 10−35 rs4385527-A 9q22.32 AOPEP Novel 3.92 8.56 x 10−05 12.03 6.16 x 10.34 4.69 x 11.99 3.89 x 0.95 10−33 10−25 10−33 rs78249997-T 10q22.2 MYOZ1 Novel −3.82 1.37 x 10−04 −12.08 8.75 x −10.32 5.80 x −12.03 2.45 x 0.98 6 10−34 10−25 10−33 rs7172038-T 15q24.1 NEO1 Novel −2.73 6.43 x 10−03 −12.58 4.78 x −10.04 1.02 x −12.37 3.76 x 0.74 10−36 10−23 10−35 rs2738413-A 14q23.2 ESR2 Novel 2.97 3.03x10−03 11.61 2.55 x 9.52 1.67 x 11.47 1.80 x 0.85 10−31 10−21 10−30 rs6891790-T 5q35.1 NKX2-5 (Malik et −4.97 6.67x10−07 −9.59 4.53 x −9.29 1.57 x −9.80 1.16 x 1.00 al.) [7] 10−22 10−20 10−22 rs2857265-A 2q31.2 FKBP7 Novel 2.76 5.85 x 10−03 10.16 4.58 x 8.42 3.65 x 10.06 8.29 x 0.70 10−24 10−17 10−24 rs10753933-T 1q32.1 PPFIA4 Novel 3.72 2.09 x 10−04 9.09 9.84 x 8.24 1.70 x 9.16 5.30 x 0.99 10−20 10−16 10−20 rs74399915-T 11q24.3 C11orf45 Novel 3.38 7.20 x 10−04 9.05 1.23 x 8.03 1.01 x 9.08 1.11 x 0.92 PRS for Cardioembolic Stroke Prediction 10−19 10−15 10−19 July 2022 | Volume 9 | Article 940696 rs13191450-A 6q22.31 HSF2 Novel 2.93 3.51 x 10−03 8.88 9.97 x 7.65 1.95 x 8.86 7.96 x 0.81 10−19 10−14 10−19 rs2834618-T 21q22.12 RUNX1 Novel 3.31 9.48 x 10−04 8.43 3.41x 7.56 3.96 x 8.47 2.37 x 0.96 10−17 10−14 10−17 rs56181519-T 2q31.1 WIPF1 Novel −2.73 6.31 x 10−03 −8.60 6.46 x −7.35 1.98 x −8.56 1.11 x 0.82 10−18 10−13 10−17 rs12908004-A 15q25.1 ARNT2 Novel −3.28 1.03 x 10−03 −8.13 4.12 x −7.35 2.03 x −8.19 2.65 x 0.96 10−16 10−13 10−16 (Continued) Frontiers in Cardiovascular Medicine | www.frontiersin.org Cárcel-Márquez et al. TABLE 1 | Continued MEGASTROKE CES AF-2018 MTAG-CES MTAG-AF SNV Locus Gene Novelty Z P Z P Z P Z P PPA rs3176326-A 6p21.2 CDKN1A Novel −3.98 6.66 x 10−05 −7.36 1.42 x −7.23 5.01 x −7.54 4.59 x 1.00 10−13 10−13 10−14 rs6747542-T 2p13.3 GMCL1 Novel 2.61 8.86 x 10−03 8.27 1.10 x 7.06 1.63 x 8.23 1.82 x 0.75 10−16 10−12 10−16 rs337705-T 5q22.3 KCNN2 Novel −2.56 1.04 x 10−02 −8.29 1.63 x −7.05 1.77 x −8.25 1.57 x 0.72 10−16 10−12 10−16 rs41292535-A 1p32.2 EPS15 Novel 3.81 1.39 x 10−04 7.16 7.42 x 6.99 2.73 x 7.33 2.34 x 0.96 10−13 10−12 10−13 rs140185678-A 16p13.3 RPL3L Novel 2.92 3.54 x 10−03 7.61 2.43 x 6.79 1.13 x 7.64 2.13 x 0.89 10−14 10−11 10−14 rs76774446-A 17q21.32 GOSR2 Novel 4.38 1.20 x 10−05 6.15 8.72 6.63 3.32 x 6.43 1.25 x 0.99 x10−10 10−11 10−10 rs55754224-T 4q26 CAMK2D Novel 2.80 5.05 x 10−03 7.39 2.15 x 6.57 4.92 x 7.41 1.22 x 0.80 10−13 10−11 10−13 rs79187193-A 1q21.2 GJA5 Novel −2.37 1.78 x 10−02 −7.59 3.15 x −6.47 9.79 x −7.56 4.08 x 0.71 10−14 10−11 10−14 rs12260801-T 10q25.2 PDCD4 Novel 4.56 5.22 x 10−06 5.53 2.94 x 6.31 2.79 x 5.86 4.62 x 1.00 10−08 10−10 10−09 7 rs2269001-A 7q36.1 KCNH2 Novel −2.89 3.94 x 10−03 −6.63 4.01 x −6.11 1.00 x −6.70 2.06 x 0.73 10−11 10−09 10−11 rs6598541-A 15q26.3 IGF1R Novel 2.69 7.17 x 10−03 6.35 2.22 x 5.81 6.21 x 6.41 1.48 x 0.72 10−10 10−09 10−10 rs11125871-T 2p15 C2orf74 Novel −3.24 1.17 x 10−03 −5.79 6.42 x −5.75 9.19 x −5.95 2.71 x 0.87 10−09 10−09 10−09 rs635634-T 9q34.2 ABO (Williams 5.10 3.31 x 10−07 4.20 2.74 x 5.72 1.04 x 4.66 3.13 x 1.00 et al.) 10−05 10−08 10−06 [10]; Malik et al. [7] rs10272350-A 7q11.23 TMEM60 Novel 4.53 6.13 x 10−06 4.62 3.41 x 5.68 1.32 x 4.99 5.93 x 0.99 10−06 10−08 10−07 PRS for Cardioembolic Stroke Prediction rs116600817-A 17p13.1 TNFSF12 Novel 2.88 3.92 x 10−03 6.00 2.45 x 5.68 1.33 x 6.10 1.07 x 0.85 July 2022 | Volume 9 | Article 940696 10−09 10−08 10−09 rs13010313-T 2q22.3 ZEB2 Novel 3.52 4.21 x 10−04 5.45 5.72 x 5.67 1.41 x 5.65 1.57 x 0.90 10−08 10−08 10−08 rs2885697-T 1p34.2 SCMH1 Novel −2.54 1.09 x 10−02 −6.27 2.88 x −5.67 1.41 x −6.32 2.70 x 0.72 10−10 10−08 10−10 rs11057583-A 12q24.31 NCOR2 Novel 3.82 1.35 x 10−04 5.19 2.10 x 5.67 1.46 x 5.45 5.13 x 0.74 10−07 10−08 10−08 (Continued) Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction Biological Processes of Loci Associated With CES and AF and Biological Processes PPA 0.78 0.96 1.00 0.78 0.87 0.96 of Loci Associated Exclusively With AF The GO of biological processes from the Genome-Wide loci of the MTAG-CES analysis revealed 98 enriched gene 2.94 x 1.78 x 7.96 x 1.62 x 3.19 x 6.59 x 10−09 10−08 10−10 10−09 10−09 10−08 sets (Supplementary Figure 1, Supplementary Table 7); the top P MTAG-AF biological processes were cardiac conduction, cardiac muscle cell SNV, Index Single Nucleotide Variant and assessed allele; Z, Z-score; P, P-value. Loci highlighted in bold are the ones not previously associated with AF. PPA, posterior probability of model 3 of GWAS-PW. contraction, and cardiac muscle contraction. A biological process analysis of the genes associated −5.94 −6.15 −5.92 5.63 6.03 5.40 exclusively with AF (Supplementary Figure 2, Z Supplementary Table 8) revealed 41 biological processes exclusive to AF risk and mainly associated with cardiac development processes (Supplementary Figure 1, 1.65 x 2.06 x 2.13 x 2.38 x 3.06 x 3.50 x 10−08 10−08 10−08 10−08 10−08 10−08 Supplementary Table 9). P MTAG-CES Polygenic Risk Score The training set was composed of 1,212 CES patients and −5.64 −5.60 −5.54 5.61 5.58 5.51 4,501 controls and the test set of 303 CES patients and 1,125 Z controls from GENERACION. No significant differences in clinical variables were found between the training and the test sets (Supplementary Table 10). 5.71 x 5.06 x 1.39 x 2.96 x 5.24 x 2.30 x 10−09 10−08 10−09 10−09 10−09 10−07 For model-1, the PRS with the highest r2 in the training set P 2 (r = 0.018) was obtained with an SNV p-value threshold of AF-2018 5 × 10−8 , comprising a total of 93 SNVs (Figure 3). Age, sex, and hypertension were the only variables for which information −5.81 −6.08 −5.82 was available for >90% of the patients, and therefore, the only 5.44 5.94 5.18 Z ones considered in the multivariable model as mentioned in the Methods” section. The three variables were significantly associated and therefore included in model-2. For model-3, we 2.34 x 10−03 6.24 x 10−04 8.30 x 10−03 5.67 x1 0−03 4.47 x 10−03 3.58 x 10−04 added the PRS to model-2, and all remaining variables were MEGASTROKE CES significant (Supplementary Table 11), including the PRS with a P Z-value of 4.33 and a p-value of 1.28 × 10−5 . The AUC in the test set for the different models was 0.581 in model-1, 0.947 in model-2, and 0.950 in model-3. AUPRC was 0.271 in model-1, 0.877 in model-2, and 0.883 in model- −3.05 −2.64 −2.84 3.42 2.77 3.56 Z 3 (Figure 3). Comparing AUC, there was significantly better discrimination in model-3 than model-2 (Z-score = −2.50, p-value = 0.01). AUC and AUPRC for each individual predictor can be found in Supplementary Figure 3. Novelty Additionally, the NRI categorical and quantitative and IDI Novel Novel Novel Novel Novel Novel showed a significant reclassification when quartiles of score risk were analyzed (Table 2). KIAA1755 LRMDA DISCUSSION FILIP1 CALU Gene PTK2 FGF5 Using an MTAG with the two biggest cohorts of CES (7) and AF (6) to date, we found 44 genome-wide significant loci associated with CES. The prioritized genes of this loci were 20q11.23 10q22.3 4q21.21 8q24.3 6q14.1 7q32.1 Locus involved in biological processes such as cardiac conduction and contraction. Nevertheless, the 51 loci associated exclusively with AF (not associated with CES as shown in the GWAS-pairwise) TABLE 1 | Continued were mainly associated with cardiac development processes. This rs12211255-A rs11782313-T rs11099098-T rs55985730-T highlights the possible role in the risk of stroke due to AF of genes rs1898096-A rs3746471-A related to cardiac conduction and contraction instead of the cardiac development process and thereby would help to develop SNV more specific prevention drugs. Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction TABLE 2 | Reclassification table comparing CES models with and without PRS addition. Risk Category Age+Sex+HT+PRS model Risk Category Age+Sex+HT model No. No. No. No. % Reclassified Non cases Q1 4,062 23 0 0 1 Q2 25 174 16 0 19 Q3 0 23 120 4 18 Q4 0 0 3 50 0 Cases Q1 149 7 0 0 4 Q2 7 124 10 0 12 Q3 0 12 171 34 21 Q4 0 0 18 677 3 NRI (Categorical) [95% CI]: 0.0134 [−0.0024–0.0291]; p-value: 0.09688 NRI (Continuos) [95% CI]: 0.1416 [0.0782–0.2049]; p-value: 1e-5 IDI [95% CI]: 0.0029 [9e-04−0.0049]; p-value: 0.00431 Q, quartile; NRI, Net Reclassification Index; IDI, integrated discrimination improvement. Eleven loci significantly associated with CES were replicated increase the levels of procoagulant factors VII, IX, X, XII, and in the independent cohort. Their prioritized genes are listed XIII and reduce the concentrations of the anticoagulant factors as follows: PITX2, ZFHX3, NKX2-5, CAV1, IGF1R, KIAA1755, protein S and antithrombin (22). Therefore, it might be a stroke NEURL1, GORAB, ESR2, ZEB2, and WIPF1. Of the genes risk marker. associated with these loci, PITX2, ZFHX3, and NKX2-5 were IGF1R encodes the insulin-like growth factor (IGF) 1 receptor, already known to be associated with CES and AF. Eight were that is, the main receptor mediating IGF signaling in the heart new CES associations; seven of them were previously associated (23). Inhibition of the IGF receptor decreases the proliferation with AF, namely, CAV1, ESR2, GORAB, IGF1R, NEURL1, WIPF1, of cardiomyocytes in murine embryonic stem cells (23). ZEB2 and ZEB2; and KIAA1755 was a completely new association with encodes the zinc finger E-box-binding homeobox 2 protein that CES, not being previously associated with AF. regulates cardiac fibroblast activation. An aberrant activation One could think that by increasing the statistical power to find could lead to structural changes prone to develop AF. CES-associated SNVs through enrichment of AF patients, part KIAA1755 has not previously been found associated with of the associations is due to actually being associated only with AF. The index variant of this locus, rs3746471-A, encodes AF. For this reason, we ensure that SNVs belonged to genomic for R1045W amino acid change, predicted to be deleterious regions associated with AF and CES through GWAS-pairwise according to SIFT. rs3746471-A has been previously described (PPA-3 > 0.6). Therefore, these 11 SNVs could be markers of as associated with heart rate (24–26) and PR interval (26) and stroke risk among patients with ESUS or among AF patients, as is remarkably suggestively associated with stroke infarct volume they are SNVs located in genomic regions that are not exclusively (p-value = 6.80 × 10−7 ) (27, 28). KIAA1755 is predicted to associated with either CES or FA, but with both. encode an uncharacterized protein and is only characterized at Of the new loci associations with CES, we could highlight the transcriptional level. The transcript is highly expressed in the some genes. CAV1 encodes caveolin-1, the principal structural brain and nerves and is also expressed in the heart. component of caveolae organelles in smooth muscle cells and We also found a novel locus suggestive to be associated with endothelial cells (17). Caveolin-1 confers an anti-AF effect by AF: 6q14.1, being FILIP1 the prioritized gene linked with the mediating atrial structural remodeling through its antifibrotic leading SNV of the locus. This gene encodes a filamin A binding action (18). Also, it plays a key role in how gas6 exerts its protein and has been identified as a regulator of myogenesis prothrombotic role in the vasculature (19). Genetic disruption differentiation in human cells and in an in vivo mouse model of caveolin-1 in mice induces a severe biventricular hypertrophy (29). In the replication stage, this SNV was found suggestive (p- with systolic and diastolic heart failure (20). That supports the value = 0.09), highly probable due to the small sample size in relevance that caveolin-1 might have in other causes of CES comparison with MTAG analysis. as symptomatic congestive heart failure with reduced ejection The PRS generated with the SNVs from MTAG-CES was fraction (21), or its importance in ESUS as a marker of an occult associated with CES independently of age, sex, and hypertension, FA or left ventricular disfunction, which could benefit from being simpler than other PRS that needs a major number of SNVs anticoagulant treatment. for association (30). We found that the addition of our PRS to a ESR2 encodes for the estrogen receptor beta, one of the model with age, sex, and hypertension significantly improves the receptors that mediates the biological effects of estrogens, which discriminatory power to detect CES. Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction FIGURE 3 | Polygenic risk score (PRS) performance. (A) is a bar plot of the r 2 for the PRS models of eight different thresholds in the training set. (B) represents the p-value variation along the full range of thresholds evaluated in the training set. (C) shows ROC curves, and (D) shows precision-recall curves for the PRS performance in the independent test set. AUC, area under the ROC curve; AUPRC, area under the precision-recall curve; HT, hypertension. Interestingly, the quantitative NRI was estimated in 14.16%, effect-size estimates for the first trait and increasing the false which is the proportion of cases correctly assigned to a higher discovery rate (and inflated type I error rate) (8). Nevertheless, probability of CES, among controls correctly assigned to a lower MTAG estimation of χ2 revealed a scenario expected to be strong probability by an updated model adding our PRS compared with against false positives, as tested in the original publication (8), the initial model without it. and less evidence on genomic inflation was observed. Besides, As limitations, the difference in the sample size between the we used GWAS-pairwise to ensure that the novel loci were not two original studies could lead to significant results for SNVs associated with only one of the traits, but with both at the same that are truly null for one trait but not for another, biasing time, having a PPA-3 > 0.6. But even more important, as usually Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction in this kind of studies, we validated the significant loci found in FUNDING this MTAG-CES and MTAG-AF in a GWAS of an independent European cohort. The small size of this last cohort underpowers J. Cárcel-Márquez has received funding through an AGAUR the ability to find significant results. However, we were able to Contract (Agència de Gestió d’Ajuts Universitaris i de Recerca; replicate 11 leading SNVs from the total number of significant FI_DGR 2019, grant number 2020FI_B1 00157) co-financed loci in the MTAG-CES and suggest one new potential locus in with Fons Social Europeu (FSE) (https://agaur.gencat.cat). From the MTAG-AF. Instituto de Salud Carlos III: E. Muiño is funded by a Another limitation is that we have only found loci associated Río Hortega Contract (CM18/00198), M. Lledós is funded with CES risk due to AF. Therefore, further multitrait analysis by a PFIS Contract (Contratos Predoctorales de Formación should be performed with different traits to uncover the different en Investigación en Salud, FI19/00309), C. Gallego-Fabrega high-risk sources of CES. Nevertheless, our aim was to better is supported by a Sara Borrell Contract (CD20/00043) and characterize patients with CES due to AF as it is the most frequent Fondo Europeo de Desarrollo Regional (ISCIII-FEDER), T. cause of this type of stroke, for subsequently being able to find Sobrino (CPII17/00027), and F. Campos (CPII19/00020) are tools to detect those patients with a higher risk of developing recipients of research contracts from the Miguel Servet a stroke due to an occult AF among ESUS for guiding future Program (https://www.isciii.es). This study has been funded by clinical trials with anticoagulant therapy. Instituto de Salud Carlos III PI15/01978, PI17/02089, PI18- In conclusion, we found and replicate 11 loci associated with 01338, and RICORS-ICTUS RD21/0006/0006 (Instituto de CES, with eight of them having new associations. We showed that Salud Carlos III), by Marató TV3 support of the Epigenesis their leading SNVs are in genomic regions related to both, CES study (https://www.ccma.cat/tv3/marato/), by the Fundació and AF, suggesting that they, together with the creation of a PRS Docència i Recerca FMT grant for the Epigenesis project that improves the predictive models of CES, might allow to better (https://www.mutuaterrassa.com), by Eranet-Neuron of the stratify the risk of stroke and its possible etiology to guide future Ibiostroke project (AC19/00106) (https://www.neuron-eranet. clinical trials of anticoagulant therapy in AF or ESUS patients for eu), by Boehringer Ingelheim of the SEDMAN Study (https:// a personalized medicine. www.boehringer-ingelheim.it), and GCAT Cession Research Project PI-2018-01 (http://www.gcatbiobank.org). GCAT was DATA AVAILABILITY STATEMENT funded by Acción de Dinamización del ISCIII-MINECO and the Ministry of Health of the Generalitat of Catalunya (ADE The data that supports the findings of this study are available 10/00026); and have additional suport by the Agència de from the corresponding author upon reasonable request. Gestió d’Ajuts Universitaris i de Recerca (AGAUR) (2017- SGR 529). ETHICS STATEMENT ACKNOWLEDGMENTS The studies involving human participants were reviewed and approved by Comité ético de la Fundación Docència I Recerca The Genotype-Tissue Expression (GTEx) Project was funded by Mútua Terrassa. The patients/participants provided their written the Common Fund of the Office of the Director of the National informed consent to participate in this study. Written informed Institutes of Health and by NCI, NHGRI, NHLBI, NIDA, NIMH, consent was obtained from the individual(s) for the publication and NINDS. The data used for the analyses described in this of any potentially identifiable images or data included in manuscript were obtained from the GTEx Portal on 03/30/20. this article. This study uses data generated by the GCAT, Genomes for Life. Cohort study of the Genomes of Catalonia, IGTP. A full list of the investigators who contributed to the generation of the AUTHOR CONTRIBUTIONS data is available from http://www.genomesforlife.com/. IGTP Conception and design of the work: JC-M, EM, and IF-C. Data is part of the CERCA Program/Generalitat de Catalunya. This acquisition: TS, FC, JC, JA, VO, LL-C, MF, JÁ-S, CM, MR, JJ-C, study was carried out using anonymized data provided by the JR, LM-N, EL-C, RD-N, CV-B, GS-H, TS, LI, LH, PD, JK, RD, Catalan Agency for Quality and Health Assessment, within the RD-M, LP-S, PC-R, NB, LS, RdC, JM, CC, J-ML, JM-F, and IF- framework of the PADRIS Program. C. Formal analysis and methodology: JC-M, EM, CG-F, NC, ML, and IF-C. Interpretation and supervision and writing—original SUPPLEMENTARY MATERIAL draft: JC-M, EM, CG-F, NC, ML, JM, CC, J-ML, JM-F, and IF-C. All authors have been involved in drafting the article or revising The Supplementary Material for this article can be found it critically for intellectual content, writing—review and editing, online at: https://www.frontiersin.org/articles/10.3389/fcvm. and approved the submitted version. 2022.940696/full#supplementary-material Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 July 2022 | Volume 9 | Article 940696 Cárcel-Márquez et al. PRS for Cardioembolic Stroke Prediction REFERENCES caveolin-1-enriched microdomains. J Thromb Haemost. (2014) 12:395–408. doi: 10.1111/jth.12481 1. Hart RG, Diener HC, Coutts SB, Easton JD, Granger CB, O’Donnell MJ, 20. Wunderlich C, Schober K, Lange SA, Drab M, Braun-Dullaeus RC, Kasper M, et al. Embolic strokes of undetermined source: the case for a new clinical et al. Disruption of caveolin-1 leads to enhanced nitrosative stress and severe construct. Lancet Neurol. (2014) 13:429–38. doi: 10.1016/S1474-4422(13) systolic and diastolic heart failure. Biochem Biophys Res Commun. (2006) 70310-7 340:702–8. doi: 10.1016/j.bbrc.2005.12.058 2. Ntaios G. Embolic stroke of undetermined source: JACC review topic of 21. Ay H, Benner T, Arsava EM, Furie KL, Singhal AB, Matt B. et al. the week. J Am Coll Cardiol. (2020) 75:333–40. doi: 10.1016/j.jacc.2019. A computerized algorithm for etiologic classification of ischemic stroke: 11.024 the causative classification of stroke system. Stroke. (2007) 38:2979– 3. Diener H-C, Sacco RL, Easton JD, Granger CB, Bernstein RA, Uchiyama S, et 84. doi: 10.1161/STROKEAHA.107.490896 al. Dabigatran for prevention of stroke after embolic stroke of undetermined 22. Aléssio AM, Höehr NF, Siqueira LH, Ozelo MC, de Pádua Mansur A, source. N Engl J Med. (2019) 380:1906–17. doi: 10.1056/NEJMoa1813959 Annichino-Bizzacchi JM. Association between estrogen receptor alpha and 4. Flint AC, Banki NM, Ren X, Rao VA, Go AS. Detection of paroxysmal atrial beta gene polymorphisms and deep vein thrombosis. Thromb Res. (2007) fibrillation by 30-day event monitoring in cryptogenic ischemic stroke: the 120:639–45. doi: 10.1016/j.thromres.2006.10.019 stroke and monitoring for PAF in real time (SMART) registry. Stroke. (2012) 23. Díaz Del Moral S, Benaouicha M, Muñoz-Chápuli R, Carmona R. The 43:2788–90. doi: 10.1161/STROKEAHA.112.665844 insulin-like growth factor signalling pathway in cardiac development and 5. Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, regeneration. Int J Mol Sci. (2021) 23:234. doi: 10.3390/ijms23010234 et al. 2016 ESC guidelines for the management of atrial fibrillation 24. Nolte IM, Munoz ML, Tragante V, Amare AT, Jansen R, Vaez A, et al. Genetic developed in collaboration with EACTS. Eur Heart J. (2016) 37:2893– loci associated with heart rate variability and their effects on cardiac disease 962. doi: 10.1093/eurheartj/ehw210 risk. Nat Commun. (2017) 8:15805. doi: 10.1038/ncomms15805 6. Nielsen JB, Thorolfsdottir RB, Fritsche LG, Zhou W, Skov MW, Graham SE, et 25. van den Berg ME, Warren HR, Cabrera CP, Verweij N, Mifsud B, Haessler J. et al. Biobank-driven genomic discovery yields new insight into atrial fibrillation al. Discovery of novel heart rate-associated loci using the Exome Chip. Hum biology. Nat Genet. (2018) 50:1234–9. doi: 10.1038/s41588-018-0171-3 Mol Genet. (2017) 26:2346–63. doi: 10.1093/hmg/ddx113 7. Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, et 26. Common Metabolic Diseases Knowledge Portal. 2021 Jun 17. Available online al. Multiancestry genome-wide association study of 520,000 subjects identifies at: https://hugeamp.org/variant.html?variant=rs3746471 (accessed June 17, 32 loci associated with stroke and stroke subtypes. Nat Genet. (2018) 50:524– 2021) 37. doi: 10.1038/s41588–018–0058–3 27. Common Metabolic Diseases Knowledge Portal. Available online at: https:// 8. Turley P, Walters RK, Maghzian O, Okbay A, Lee JJ, Fontana MA, et al. Multi- hugeamp.org/gene.html?gene=KIAA1755 (accessed June 17, 2021). trait analysis of genome-wide association summary statistics using MTAG. 28. Pirruccello JP, Bick A, Wang M, Chaffin M, Friedman S, Yao J, et al. Nat Genet. (2018) 50:229–37. doi: 10.1038/s41588-017-0009-4 Analysis of cardiac magnetic resonance imaging in 36,000 individuals 9. Pickrell JK, Berisa T, Liu JZ, Ségurel L, Tung JY, Hinds DA. Detection and yields genetic insights into dilated cardiomyopathy. Nat Commun. (2020) interpretation of shared genetic influences on 42 human traits. Nat Genet. 11:2254. doi: 10.1038/s41467-020-15823-7 (2016) 48:709–17. doi: 10.1038/ng.3570 29. Militello G, Hosen MR, Ponomareva Y, Gellert P, Weirick T, John D, et al. A 10. Williams FMK, Carter AM, Hysi PG, Surdulescu G, Hodgkiss D, Soranzo N, novel long non-coding RNA Myolinc regulates myogenesis through TDP-43 et al. Ischemic stroke is associated with the ABO locus: the EuroCLOT study. and Filip1. J Mol Cell Biol. (2018) 10:102–17. doi: 10.1093/jmcb/mjy025 Ann Neurol. (2013) 73:16–31. doi: 10.1002/ana.23838 30. Pulit SL, Weng LC, McArdle PF, Trinquart L, Choi SH, 11. von Berg J, van der Laan SW, McArdle PF, Malik R, Kittner SJ, Mitchell Mitchell BD, et al. Atrial fibrillation genetic risk differentiates BD, et al. Alternate approach to stroke phenotyping identifies a genetic cardioembolic stroke from other stroke subtypes. Neurol Genet. (2018) risk locus for small vessel stroke. Eur J Hum Genet. (2020) 28:963– 4:1–8. doi: 10.1212/NXG.0000000000000293 72. doi: 10.1038/s41431-020-0580-5 12. Das S, Forer L, Schönherr S, Sidore C, Locke AE, Kwong A, et al. Next- Conflict of Interest: The authors declare that the research was conducted in the generation genotype imputation service and methods. Nat Genet. (2016) absence of any commercial or financial relationships that could be construed as a 48:1284–7. doi: 10.1038/ng.3656 potential conflict of interest. 13. Jiang L, Zheng Z, Qi T, Kemper KE, Wray NR, Visscher PM, et al. A resource- efficient tool for mixed model association analysis of large-scale data. Nat Publisher’s Note: All claims expressed in this article are solely those of the authors Genet. (2019) 51:1749–55. doi: 10.1038/s41588-019-0530-8 and do not necessarily represent those of their affiliated organizations, or those of 14. Mountjoy E, Schmidt EM, Carmona M, Schwartzentruber J, Peat G, Miranda the publisher, the editors and the reviewers. Any product that may be evaluated in A, et al. An open approach to systematically prioritize causal variants and this article, or claim that may be made by its manufacturer, is not guaranteed or genes at all published human GWAS trait-associated loci. Nat Genet. (2021) 53:1527–33. doi: 10.1038/s41588-021-00945-5 endorsed by the publisher. 15. Choi SW, O’Reilly PF. PRSice-2: polygenic risk score software for biobank- scale data. Gigascience. (2019) 8:1–6. doi: 10.1093/gigascience/giz082 Copyright © 2022 Cárcel-Márquez, Muiño, Gallego-Fabrega, Cullell, Lledós, 16. Bennett DA. How can I deal with missing data in my study? Aust NZJ Public Llucià-Carol, Sobrino, Campos, Castillo, Freijo, Arenillas, Obach, Álvarez-Sabín, Health. (2001) 25:464–9. doi: 10.1111/j.1467-842X.2001.tb00294.x Molina, Ribó, Jiménez-Conde, Roquer, Muñoz-Narbona, Lopez-Cancio, Millán, 17. Murata T, Lin MI, Huang Y, Yu J, Bauer PM, Giordano FJ, et al. Reexpression Diaz-Navarro, Vives-Bauza, Serrano-Heras, Segura, Ibañez, Heitsch, Delgado, Dhar, of caveolin-1 in endothelium rescues the vascular, cardiac, and pulmonary Krupinski, Delgado-Mederos, Prats-Sánchez, Camps-Renom, Blay, Sumoy, de Cid, defects in global caveolin-1 knockout mice. J Exp Med. (2007) 204:2373– Montaner, Cruchaga, Lee, Martí-Fàbregas and Férnandez-Cadenas. This is an open- 82. doi: 10.1084/jem.20062340 access article distributed under the terms of the Creative Commons Attribution 18. Yi SL, Liu XJ, Zhong JQ, Zhang Y. Role of caveolin-1 in atrial fibrillation as License (CC BY). The use, distribution or reproduction in other forums is permitted, an anti-fibrotic signaling molecule in human atrial fibroblasts. PLoS ONE. provided the original author(s) and the copyright owner(s) are credited and that the (2014)9:e85144. doi: 10.1371/journal.pone.0085144 original publication in this journal is cited, in accordance with accepted academic 19. Laurance S, Aghourian MN, Jiva Lila Z, Lemarié CA, Blostein MD. Gas6- practice. No use, distribution or reproduction is permitted which does not comply induced tissue factor expression in endothelial cells is mediated through with these terms. Frontiers in Cardiovascular Medicine | www.frontiersin.org 12 July 2022 | Volume 9 | Article 940696