Human Molecular Genetics, 2022, Vol. 31, 23, 3945–3966 https://doi.org/10.1093/hmg/ddac158 Advance access publication date 15 July 2022 Original Article Detailed stratified GWAS analysis for severe COVID-19 in four European populations 1,†, Frauke Degenhardt *, David Ellinghaus1,2,† , Simonas Juzenas1,3,† , Jon Lerga-Jaso4,† , Mareike Wendorff1,† , Douglas Maya-Miles5,6,7,† , Florian Uellendahl-Werth1 , Hesham ElAbd1 , Malte C. Rühlemann1,8 , Jatin Arora9,10,11,12,13 , Onur Özer14,15 , Ole Bernt Lenning16,17 , Ronny Myhre18 , May Sissel Vadla17,19 , Eike M. Wacker1 , Lars Wienbrandt1 , Aaron Blandino Ortiz20 , Adolfo de Salazar21,22 , Adolfo Garrido Chercoles23 , Adriana Palom24,25 , Agustín Ruiz26,27 , Alba-Estela Garcia-Fernandez28 , Albert Blanco-Grau28 , Alberto Mantovani29,30 , Alberto Zanella31,32 , Aleksander Rygh Holten33,34 , Alena Mayer35 , Alessandra Bandera31,32 , Alessandro Cherubini32 , Alessandro Protti29,30 , Alessio Aghemo29,30 , Alessio Gerussi36,37 , Alfredo Ramirez38,39,40,41,42 , Alice Braun35 , Almut Nebel1 , Ana Barreira25 , Ana Lleo29,30 , Ana Teles14,15 , Anders Benjamin Kildal 43 , Andrea Biondi44 , Andrea Caballero-Garralda45 , Andrea Ganna46 , Andrea Gori32,47 , Andreas Glück48 , Andreas Lind49 , Anja Tanck1 , Anke Hinney50 , Anna Carreras Nolla51 , Anna Ludovica Fracanzani31,32 , Anna Peschuck1 , Annalisa Cavallero52 , Anne Ma Dyrhol-Riise34,53 , Antonella Ruello54 , Antonio Julià 24 , Antonio Muscatello32 , Antonio Pesenti31,32 , Antonio Voza29,30 , Ariadna Rando-Segura55,56 , Aurora Solier57,58 , Axel Schmidt59 , Beatriz Cortes51 , Beatriz Mateos6,60 , Beatriz Nafria-Jimenez23 , Benedikt Schaefer61,62 , Björn Jensen63 , Carla Bellinghausen64 , Carlo Maj65 , Carlos Ferrando66,67 , Carmen de la Horra5,7,68,69,70 , Carmen Quereda71 , Carsten Skurk35 , Charlotte Thibeault35 , Chiara Scollo72 , Christian Herr73 , Christoph D. Spinner74 , Christoph Gassner1,75 , Christoph Lange76,77,78 , Cinzia Hu32 , Cinzia Paccapelo32 , Clara Lehmann80,81,82 , Claudio Angelini83 , Claudio Cappadona30 , Clinton Azuure14,15 , COVICAT study group, Aachen Study (COVAS), Cristiana Bianco32 , Cristina Cea28 , Cristina Sancho84 , Dag Arne Lihaug Hoff85,86 , Daniela Galimberti31,32 , Daniele Prati32 , David Haschka87 , David Jiménez57,58 , David Pestaña88 , David Toapanta67 , Eduardo Muñiz-Diaz89 , Elena Azzolini29,30 , Elena Sandoval67 , Eleonora Binatti36,37 , Elio Scarpini31,32 , Elisa T. Helbig35 , Elisabetta Casalone90 , Eloisa Urrechaga91,92 , Elvezia Maria Paraboschi 29,30 , Emanuele Pontali93 , Enric Reverter67 , Enrique J. Calderón5,7,68,69,70 , Enrique Navas71 , Erik Solligård94,95 , Ernesto Contro96 , Eunate Arana-Arri97 , Fátima Aziz67 , Federico Garcia21,22,98 , Félix García Sánchez99 , Ferruccio Ceriotti32 , Filippo Martinelli-Boneschi100,101 , Flora Peyvandi102,103 , Florian Kurth35,104,105 , Francesco Blasi106,107 , Francesco Malvestiti31 , Francisco J. Medrano7,68,69,70,108 , Francisco Mesonero6,60 , Francisco Rodriguez-Frias6,24,28,55,56,109 , Frank Hanses110,111 , Fredrik Müller34,49 , Georg Hemmrich-Stanisak1 , Giacomo Bellani112,113 , Giacomo Grasselli31,32 , Gianni Pezzoli114 , Giorgio Costantino31,32 , Giovanni Albano54 , Giulia Cardamone30 , Giuseppe Bellelli113,115 , Giuseppe Citerio113,116 , Giuseppe Foti112,113 , Giuseppe Lamorte32 , Giuseppe Matullo90 , Guido Baselli32 , Hayato Kurihara83 , Holger Neb117 , Ilaria My29 , Ingo Kurth118 , Isabel Hernández26,27 , Isabell Pink119 , Itziar de Rojas26,27 , Iván Galván-Femenia51 , Jan Cato Holter34,49 , Jan Egil Afset84,120 , Jan Heyckendorf76,77,78 , Jan Kässens1 , Jan Kristian Damås121,122 , Jan Rybniker80,81,123 , Janine Altmüller124 , Javier Ampuero5,6,7,68,125 , Javier Martín126 , Jeanette Erdmann127,128,129 , Jesus M. Banales6,130,131 , Joan Ramon Badia132 , Joaquin Dopazo 7,133 , Jochen Schneider73 , Jonas Bergan134 , Jordi Barretina135 , Jörn Walter136 , Jose Hernández Quero21,137 , Josune Goikoetxea138 , Juan Delgado5,7,68,69,70 , Juan M. Guerrero5,7,68 , Julia Fazaal59 , Julia Kraft35 , Julia Schröder59 , Kari Risnes122,139 , Karina Banasik2 , Karl Erik Müller140 , Karoline I. Gaede141,142,143 , Koldo Garcia-Etxebarria6,130 , Kristian Tonby34,53 , Lars Heggelund140,144 , Laura Izquierdo-Sanchez6,130,145 , Laura Rachele Bettini44 , Lauro Sumoy135 , Leif Erik Sander35 , Lena J. Lippert35 , Leonardo Terranova32 , Lindokuhle Nkambule146,147 , Lisa Knopp63 , Lise Tuset Gustad94,148 , Lucia Garbarino149 , Luigi Santoro32 , Luis Téllez6,60 , Luisa Roade6,25,57 , Mahnoosh Ostadreza32 , Maider Intxausti84 , Manolis Kogevinas69,150,151,152 , Mar Riveiro-Barciela6,25,56 , Marc M. Berger153 , Marco Schaefer154 , Mari E.K. Niemi46 , María A. Gutiérrez-Stampa155 , Maria Carrabba79 , Maria E. Figuera Basso1 , Maria Grazia Valsecchi156 , María Hernandez-Tejero67 , Received: January 14, 2022. Revised: June 11, 2022. Accepted: July 26, 2022 © The Author(s) 2022. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/ by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 3946 | Human Molecular Genetics, 2022, Vol. 31, No. 23 Maria J.G.T. Vehreschild157 , Maria Manunta 32 , Marialbert Acosta-Herrera 126 , Mariella D’Angiò44 , Marina Baldini32 , Marina Cazzaniga158 , Marit M. Grimsrud34,159,160 , Markus Cornberg161 , Markus M. Nöthen59 , Marta Marquié 26,27 , Massimo Castoldi54 , Mattia Cordioli46 , Maurizio Cecconi29,30 , Mauro D’Amato131,162,163 , Max Augustin80,81,82 , Melissa Tomasi32 , Mercè Boada26,27 , Michael Dreher164 , Michael J. Seilmaier165 , Michael Joannidis166 , Michael Wittig1 , Michela Mazzocco149 , Michele Ciccarelli83 , Miguel Rodríguez-Gandía6,60 , Monica Bocciolone83 , Monica Miozzo31,79 , Natale Imaz Ayo97 , Natalia Blay 51 , 22 31,32 1,167 168 169 170 Natalia Chueca , Nicola Montano , Nicole Braun , Nicole Ludwig , Nikolaus Marx , Nilda Martínez , Norwegian SARS-CoV-2 Study group, Oliver A. Cornely40,80,82,171 , Oliver Witzke172 , Orazio Palmieri173 , Pa Study Group, Paola Faverio113,174 , Paoletta Preatoni83 , Paolo Bonfanti113,175 , Paolo Omodei83 , Paolo Tentorio29 , Pedro Castro67 , Pedro M. Rodrigues6,130,131,145 , Pedro Pablo España91 , Per Hoffmann59 , Philip Rosenstiel1 , Philipp Schommers80,81,82 , Phillip Suwalski35 , Raúl de Pablo20 , Ricard Ferrer176 , Robert Bals73 , Roberta Gualtierotti31,32 , Rocío Gallego-Durán5,6,7 , Rosa Nieto57,58 , Rossana Carpani32 , Rubén Morilla5,7,68,69,70 , Salvatore Badalamenti29 , Sammra Haider177 , Sandra Ciesek178,179 , Sandra May1 , Sara Bombace29,30 , Sara Marsal24 , Sara Pigazzini46 , Sebastian Klein166 , Serena Pelusi31,32 , Sibylle Wilf ling111,180,181 , Silvano Bosari31,32 , Sonja Volland182 , Søren Brunak2 , Soumya Raychaudhuri9,10,11,12,13,183 , Stefan Schreiber1,48 , Stefanie Heilmann-Heimbach59 , Stefano Aliberti29,30 , Stephan Ripke35 , Susanne Dudman 34,49 , Tanja Wesse1 , Tenghao Zheng184 , The STORM Study group, The Humanitas Task Force, The Humanitas Gavazzeni Task Force, Thomas Bahmer48 , Thomas Eggermann118 , Thomas Illig182 , Thorsten Brenner153 , Tomas Pumarola185,186 , Torsten Feldt63 , Trine Folseraas34,159,160,187 , Trinidad Gonzalez Cejudo188 , Ulf Landmesser189 , Ulrike Protzer190,191 , Ute Hehr180 , Valeria Rimoldi30 , Valter Monzani32 , Vegard Skogen192,193 , Verena Keitel63 , Verena Kopfnagel182 , Vicente Friaza5,7,68,69,70 , Victor Andrade38,39 , Victor Moreno69,194,195,196 , Wolfgang Albrecht1 , Wolfgang Peter154,197 , Wolfgang Poller35 , Xavier Farre51 , Xiaoli Yi1 , Xiaomin Wang35 , Yascha Khodamoradi157 , Zehra Karadeniz35 , Anna Latiano173 , Siegfried Goerg198 , Petra Bacher1,199 , Philipp Koehler40,80,81 , Florian Tran1,48 , Heinz Zoller61,62 , Eva C. Schulte190,200,201 , Bettina Heidecker35 , 59 Kerstin U. Ludwig , Javier Fernández67,202 , Manuel Romero-Gómez5,6,7,68,125 , Agustín Albillos6,60 , Pietro Invernizzi36,37 , Maria Buti6,25,56 , Stefano Duga29,30 , Luis Bujanda6,130 , Johannes R. Hov34,159,160,187,‡ , Tobias L. Lenz14,15,‡ , Rosanna Asselta29,30,‡ , Rafael de Cid51,‡ , Luca Valenti31,32,‡ , Tom H. Karlsen34,159,160,187,‡ , Mario Cáceres4,203,‡ and Andre Franke1,204,‡, * 1 Institute of Clinical Molecular Biology, Christian-Albrechts-University, Kiel, Germany 2 Novo Nordisk Foundation Center for Protein Research, Disease Systems Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark 3 Life Science Centre, Institute of Biotechnology, Vilnius University, Vilnius, Lithuania 4 Institut de Biotecnologia i de Biomedicina, Universitat Autònoma de Barcelona, Barcelona, Spain 5 Hospital Universitario Virgen del Rocío de Sevilla, Sevilla, Spain 6 Centro de Investigación Biomédica en Red en Enfermedades Hepáticas y Digestivas (CIBEREHD), Instituto de Salud Carlos III (ISCIII), Madrid, Spain 7 Instituto de Biomedicina de Sevilla (IBIS), Sevilla, Spain 8 Institute for Medical Microbiology and Hospital Epidemiology, Hannover Medical School, Hannover, Germany 9 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA 10 Division of Rheumatology, Inflammation and Immunity, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA 11 Division of Genetics, Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA 12 Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA 13 Center for Data Sciences, Brigham and Women’s Hospital, Boston, MA, USA 14 Research Group for Evolutionary Immunogenomics, Max Planck Institute for Evolutionary Biology, Plön, Germany 15 Research Unit for Evolutionary Immunogenomics, Department of Biology, University of Hamburg, Hamburg, Germany 16 Research Department, Stavanger University Hospital, Stavanger, Norway 17 Randaberg Municipality, Stavanger, Norway 18 Division of Health Data and Digitalization, Department of Genetics and Bioinformatics (HDGB), Norwegian Institute of Public Health, Oslo, Norway 19 Department of Quality and Health Technology, Faculty of Health Sciences, University of Stavanger, Stavanger, Norway 20 Department of Intensive Care, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), University of Alcalá, Madrid, Spain 21 Ibs.Granada Instituto de Investigación Biosanitaria, Granada, Spain 22 Microbiology Unit. Hospital Universitario Clinico San Cecilio, Granada, Spain 23 Clinical Biochemistry Department, Osakidetza Basque Health Service, Donostialdea Integrated Health Organisation, San Sebastian, Spain 24 Vall d’Hebron Institut de Recerca (VHIR), Vall d’Hebron Hospital Universitari, Barcelona, Spain 25 Liver Unit, Department of Internal Medicine, Hospital Universitari Vall d’Hebron, Vall d’Hebron Barcelona Hospital Campus, Barcelona, Spain 26 Research Center and Memory Clinic, Ace Alzheimer Center Barcelona, Universitat Internacional de Catalunya, Barcelona, Spain 27 CIBERNED, Network Center for Biomedical Research in Neurodegenerative Diseases, National Institute of Health Carlos III, Madrid, Spain 28 Department of Biochemistry, University Hospital Vall d’Hebron, Barcelona, Spain 29 IRCCS Humanitas Research Hospital, Milan, Italy 30 Department of Biomedical Sciences, Humanitas University, Milan, Italy 31 University of Milan, Milan, Italy 32 Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy 33 Department of Acute Medicine, Oslo University Hospital, Oslo, Norway 34 Institute of Clinical Medicine, University of Oslo, Oslo, Norway 35 Charite Universitätsmedizin Berlin, Berlin, Germany Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3947 36 European Reference Network on Hepatological Diseases (ERN RARE-LIVER), San Gerardo Hospital, Monza, Italy 37 Division of Gastroenterology, School of Medicine and Surgery, Center for Autoimmune Liver Diseases, University of Milano-Bicocca, Milan, Italy 38 Division of Neurogenetics and Molecular Psychiatry, Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany 39 Department of Neurodegenerative Diseases and Geriatric Psychiatry, University Hospital Bonn, Medical Faculty, Bonn, Germany 40 Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany 41 German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany 42 Department of Psychiatry, Glenn Biggs Institute for Alzheimer’s and Neurodegenerative Diseases, San Antonio, TX, USA 43 Department of Anesthesiology and Intensive Care, University Hospital of North Norway, Tromsø, Norway 44 Pediatric Departement, Centro Tettamanti-European Reference Network (ERN) PaedCan, EuroBloodNet, MetabERN-University of Milano-Bicocca-Fondazione MBBM/Ospedale, San Gerardo, Italy 45 Biochemistry Department, Echevarne Laboratory, Barcelona, Spain 46 Institute for Molecular Medicine Finland, University of Helsinki, Helsinki, Finland 47 Centre for Multidisciplinary Research in Health Science (MACH), University of Milan, Milan, Italy 48 Klinik für Innere Medizin I, Universitätsklinikum Schleswig-Holstein, Kiel, Germany 49 Department of Microbiology, Oslo University Hospital, Oslo, Norway 50 Department of Child and Adolescent Psychiatry, University Hospital Essen, University of Duisburg-Essen, Essen, Germany 51 Genomes for Life-GCAT Lab, Germans Trias i Pujol Research Institute (IGTP), Badalona, Spain 52 Laboratory of Microbiology, San Gerardo Hospital, Monza, Italy 53 Department of Infectious diseases, Oslo University Hospital, Oslo, Norway 54 Humanitas Gavazzeni-Castelli, Bergamo, Italy 55 Microbiology Department, Hospital Universitari Vall d’Hebron, Barcelona, Spain 56 Universitat Autònoma de Barcelona, Bellatera, Spain 57 Department of Respiratory Diseases, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), University of Alcalá, Madrid, Spain 58 Centro de Investigación Biomédica en Red en Enfermedades Respiratorias (CIBERES), University of Alcalá, Madrid, Spain 59 Institute of Human Genetics, University of Bonn School of Medicine & University Hospital Bonn, Bonn, Germany 60 Department of Gastroenterology, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), University of Alcalá, Madrid, Spain 61 Department of Medicine I, Gastroenterology, Hepatology and Endocrinology, Medical University of Innsbruck, Innsbruck, Austria 62 Christian Doppler Laboratory of Iron and Phosphate Biology at the Department of Medicine I, Medical University of Innsbruck, Innsbruck, Austria 63 Department of Gastroenterology, Hepatology and Infectious Diseases, University Hospital Duesseldorf, Medical Faculty Heinrich Heine University, Duesseldorf, Germany 64 Department of Respiratory Medicine and Allergology, University Hospital, Goethe University, Frankfurt am Main, Germany 65 Institute of Genomic Statistics and Bioinformatics, University Hospital Bonn, Medical Faculty University of Bonn, Bonn, Germany 66 Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES), Madrid, Spain 67 Hospital Clinic, University of Barcelona, and IDIBAPS, Barcelona, Spain 68 University of Sevilla, Sevilla, Spain 69 Centro de Investigación Biomédica en Red de Epidemiología y Salud Pública (CIBERESP), Madrid, Spain 70 Consejo Superior de Investigaciones Científicas, Sevilla, Spain 71 Department of Infectious Diseases, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), University of Alcalá, Madrid, Spain 72 Department of Transfusion Medicine and Haematology Laboratory, San Gerardo Hospital, Monza, Italy 73 Department of Internal Medicine V—Pneumology, Allergology and Intensive Care Medicine, University Hospital Saarland, Homburg, Germany 74 Department of Internal Medicine II, School of Medicine, University Hospital rechts der Isar, Technical University of Munich, Munich, Germany 75 Institute of Translational Medicine, Private University in the Principality of Liechtenstein (UFL), Triesen, Liechtenstein 76 Respiratory Medicine and International Health, University of Lübeck, Lübeck, Germany 77 Division of Clinical Infectious Diseases, Research Center Borstel, Borstel, Germany 78 Clinical Tuberculosis Unit, German Center for Infection Research (DZIF), Borstel, Germany 79 Internal Medicine Department, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy 80 Department I of Internal Medicine, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany 81 Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, Germany 82 German Center for Infection Research (DZIF), Cologne, Germany 83 Humanitas Clinical and Research Center, IRCCS, Milan, Italy 84 Osakidetza Basque Health Service, Basurto University Hospital, Respiratory Service, Bilbao, Spain 85 Department of Clinical and Molecular Medicine, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, Trondheim, Norway 86 Department of Medicine, Møre and Romsdal Hospital Trust, Ålesund, Norway 87 Department of Internal Medicine II, Medical University of Innsbruck, Innsbruck, Austria 88 Department of Anesthesiology and Critical Care, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), University of Alcalá, Madrid, Spain 89 Immunohematology Department, Banc de Sang i Teixits, Autonomous University of Barcelona, Barcelona, Spain 90 Department of Medical Sciences, Università degli Studi di Torino, Turin, Italy 91 Osakidetza Basque Health Service, Galdakao Hospital, Respiratory Service, Galdakao, Spain 92 Biocruces Bizkaia Health Research Institute, Barakaldo, Spain 93 Department of Infectious Diseases, E.O. Ospedali Galliera, Genova, Italy 94 Geminicenter for Sepsis Research, Institute of Circulation and Medical Imaging (ISB), NTNU, Trondheim, Norway 95 Clinic of Anesthesia and Intensive Care, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway 96 Accident and Emergency and Emergency Medicine Unit, San Gerardo Hospital, Monza, Italy 97 Biocruces Bizkaia Health Research Institute, Barakaldo, Spain 98 Centro de Investigación Biomédica en Red en Enfermedades Infecciosas (CIBERINFEC), Instituto de Salud Carlos III (ISCIII), Madrid, Spain 99 Histocompatibilidad y Biologia Molecular, Centro de Transfusion de Madrid, Madrid, Spain 100 Department of Pathophysiology and Transplantation, Dino Ferrari Center, University of Milan, Milan, Italy 101 Neurology Unit, IRCCS Fondazione Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy 102 Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy 103 Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Milan, Italy 3948 | Human Molecular Genetics, 2022, Vol. 31, No. 23 104 Department of Tropical Medicine, Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany 105 Department of Medicine I, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany 106 Respiratory Unit, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy 107 Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Milan, Italy 108 Internal Medicine Department, Virgen del Rocio University Hospital, Sevilla, Spain 109 Biochemistry Department, University Hospital Vall d’Hebron, Barcelona, Spain 110 Emergency Department, University Hospital Regensburg, Regensburg, Germany 111 Department for Infectious Diseases and Infection Control, University Hospital Regensburg, Regensburg, Germany 112 Department Emergency, Anesthesia and Intensive Care, San Gerardo Hospital, Monza, Italy 113 School of Medicine and Surgery, University of Milano-Bicocca, Milan, Italy 114 Fondazione Grigioni per il Morbo di Parkinson and Parkinson Institute, ASST Gaetano Pini-CTO, Milan, Italy 115 Acute Geriatric Unit, San Gerardo Hospital, Monza, Italy 116 Neurointensive Care Unit, San Gerardo Hospital, Monza, Italy 117 Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt, Frankfurt am Main, Germany 118 Institute of Human Genetics, Medical Faculty, RWTH Aachen University, Aachen, Germany 119 Department of Pneumology, Hannover Medical School, Hannover, Germany 120 Department of Medical Microbiology, Clinic of Laboratory Medicine, St. Olavs hospital, Trondheim, Norway 121 Department of Infectious Diseases, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway 122 Department of Clinical and Molecular Medicine, NTNU, Trondheim, Norway 123 German Centre for Infection Research (DZIF), Cologne, Germany 124 Cologne Center for Genomics (CCG), University of Cologne, Cologne, Germany 125 Digestive Diseases Unit, Virgen del Rocio University Hospital, Institute of Biomedicine of Seville, University of Seville, Seville, Spain 126 Institute of Parasitology and Biomedicine Lopez-Neyra, CSIC, Granada, Spain 127 Institute for Cardiogenetics, University of Lübeck, Lübeck, Germany 128 German Research Center for Cardiovascular Research, Lübeck, Germany 129 University Heart Center Lübeck, Lübeck, Germany 130 Department of Liver and Gastrointestinal Diseases, Biodonostia Health Research Institute—Donostia University Hospital, University of the Basque Country (UPV/EHU), CIBERehd, Ikerbasque, San Sebastian, Spain 131 Ikerbasque, Basque Foundation for Science, Bilbao, Spain 132 Respiratory ICU, Institut Clínic Respiratory, Hospital Clinic, University of Barcelona, and IDIBAPS, Barcelona, Spain 133 Bioinformatics Area, Fundación Progreso y Salud, and Institute of Biomedicine of Sevilla (IBIS), Sevilla, Spain 134 Department of Research, Ostfold Hospital Trust, Gralum, Norway 135 Germans Trias i Pujol Research Institute (IGTP), Badalona, Spain 136 Department of Genetics & Epigenetics, Saarland University, Saarbrücken, Germany 137 Department of Infectious Diseases, Hospital Universitario Clinico San Cecilio, Granada, Spain 138 Infectious Diseases Service, Osakidetza, Biocruces Bizkaia Health Research Institute, Barakaldo, Spain 139 Department of Research, St. Olav Hospital, Trondheim University Hospital, Trondheim, Norway 140 Medical Department, Drammen Hospital, Vestre Viken Hospital Trust, Drammen, Norway 141 Research Center Borstel, BioMaterialBank Nord, Borstel, Germany 142 German Center for Lung Research (DZL), Airway Research Center North (ARCN), Borstel, Germany 143 Popgen 2.0 Network (P2N), Kiel, Germany 144 Department of Clinical Science, University of Bergen, Bergen, Norway 145 Biodonostia Health Research Institute, Donostia University Hospital, San Sebastian, Spain 146 Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, MA, USA 147 Stanley Center for Psychiatric Research & Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA 148 Clinic of Medicine and Rehabilitation, Levanger Hospital, Nord-Trondelag Hospital Trust, Levanger, Norway 149 HLA Laboratory, E.O. Ospedali Galliera, Genova, Italy 150 ISGlobal, Barcelona, Spain 151 Universitat Pompeu Fabra (UPF), Barcelona, Spain 152 IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain 153 Department of Anesthesiology and Intensive Care Medicine, University Hospital Essen, University Duisburg-Essen, Essen, Germany 154 Stefan-Morsch-Stiftung, Birkenfeld, Germany 155 Osakidetza, OSI Donostialdea, Altza Primary Care, Biodonostia Health Research Institute, San Sebastián, Spain 156 Center of Bioinformatics, Biostatistics and Bioimaging, School of Medicine and Surgery, University of Milano-Bicocca, Milan, Italy 157 Department of Internal Medicine, Infectious Diseases, University Hospital Frankfurt & Goethe University Frankfurt, Frankfurt am Main, Germany 158 Phase 1 Research Centre, ASST Monza, School of Medicine and Surgery, University of Milano-Bicocca, Monza, Italy 159 Division of Surgery, Inflammatory Diseases and Transplantation, Research Institute for Internal Medicine, Oslo University Hospital Rikshospitalet and University of Oslo, Oslo, Norway 160 Division of Surgery, Inflammatory Diseases and Transplantation, Department of Transplantation Medicine, Norwegian PSC Research Center, Oslo University Hospital Rikshospitalet, Oslo, Norway 161 Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Hannover, Germany 162 Gastrointestinal Genetics Lab, CIC bioGUNE—BRTA, Derio, Spain 163 Department of Medicine and Surgery, LUM University, Casamassima, Italy 164 Department of Pneumology and Intensive Care Medicine, University Hospital Aachen, Aachen, Germany 165 Munich Clinic Schwabing, Academic Teaching Hospital, Ludwig-Maximilians-University (LMU), Munich, Germany 166 Division of Intensive Care and Emergency Medicine, Department of Internal Medicine, Medical University Innsbruck, Innsbruck, Austria 167 University Hospital Schleswig-Holstein, Kiel, Germany 168 Department of Human Genetics, Center of Human and Molecular Biology, University Hospital Saarland, Homburg/Saar, Germany 169 Department of Internal Medicine I, RWTH Aachen University Hospital, Aachen, Germany 170 Department of Anesthesiology, Hospital Universitario Ramón y Cajal, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), Madrid, Spain 171 Faculty of Medicine and University Hospital Cologne, Clinical Trials Centre Cologne (ZKS Köln), University of Cologne, Cologne, Germany 172 Department of Infectious Diseases, University Hospital Essen, University Duisburg-Essen, Essen, Germany 173 Gastroenterology Unit, Fondazione IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy 174 Pulmonary Unit, San Gerardo Hospital, Monza, Italy 175 Infectious Diseases Unit, San Gerardo Hospital, Monza, Italy 176 Intensive Care Department, Vall d’Hebron University Hospital, SODIR-VHIR research group, Barcelona, Spain Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3949 177 Department of Medicine, Møre & Romsdal Hospital Trust, Molde, Norway 178 Institute of Medical Virology, University Hospital Frankfurt, Goethe University, Frankfurt am Main, Germany 179 German Centre for Infection Research (DZIF), Frankfurt am Main, Germany 180 Zentrum für Humangenetik Regensburg, Regensburg, Germany 181 Department of Neurology, Bezirksklinikum Regensburg, University of Regensburg, Regensburg, Germany 182 Hannover Unified Biobank, Hannover Medical School, Hannover, Germany 183 Centre for Genetics and Genomics Versus Arthritis, Centre for Musculoskeletal Research, Manchester Academic Health Science Centre, The University of Manchester, Manchester, UK 184 School of Biological Sciences, Monash University, Clayton, VIC, Australia 185 Department of Microbiology, University Hospital Vall d’Hebron, Barcelona, Spain 186 Autonoma University of Barcelona, Barcelona, Spain 187 Division for Cancer Medicine, Surgery and Transplantation, Section for Gastroenterology, Department of Transplantation Medicine, Oslo University Hospital Rikshospitalet, Oslo, Norway 188 Biochemistry Unit, Hospital Universitario Clinico San Cecilio, Granada, Spain 189 Berlin Institute of Health, Charite Universitätsmedizin Berlin, Berlin Germany 190 Institute of Virology, Technical University Munich/Helmholtz Zentrum München, Munich, Germany 191 German Center for Infection Research (DZIF), Munich, Germany 192 Department of Infectious Diseases, University Hospital of North Norway, Tromsø, Norway 193 Faculty of Health Sciences, UIT The Arctic University of Norway, Norway 194 Catalan Institute of Oncology (ICO), Barcelona, Spain 195 Bellvitge Biomedical Research Institute (IDIBELL), Barcelona, Spain 196 Universitat de Barcelona (UB), Barcelona, Spain 197 Faculty of Medicine and University Hospital Cologne, Institute for Transfusion Medicine, University of Cologne, Cologne, Germany 198 Institute of Transfusionsmedicine, University Hospital Schleswig-Holstein (UKSH), Kiel, Germany 199 Institute of Immunology, Christian-Albrechts-University of Kiel and UKSH Schleswig-Holstein, Kiel, Germany 200 Institute of Psychiatric Phenomics and Genomics, University Medical Center, University of Munich, Munich, Germany 201 Department of Psychiatry, University Medical Center, University of Munich, Munich, Germany 202 European Foundation for the Study of Chronic Liver Failure (EF-CLIF), Barcelona, Spain 203 ICREA, Barcelona, Spain 204 University Hospital Schleswig-Holstein (UKSH), Kiel, Germany *To whom correspondence should be addressed. Email: f.degenhardt@ikmb.uni-kiel.de (Frauke Degenhardt) and a.franke@mucosa.de (Andre Franke) † The authors wish it to be known that, in their opinion, the first six authors should be regarded as joint First Authors. ‡ The authors wish it to be known that, in their opinion, the last seven authors should be regarded as joint Last Authors. Abstract Given the highly variable clinical phenotype of Coronavirus disease 2019 (COVID-19), a deeper analysis of the host genetic contribution to severe COVID-19 is important to improve our understanding of underlying disease mechanisms. Here, we describe an extended genome-wide association meta-analysis of a well-characterized cohort of 3255 COVID-19 patients with respiratory failure and 12 488 population controls from Italy, Spain, Norway and Germany/Austria, including stratified analyses based on age, sex and disease severity, as well as targeted analyses of chromosome Y haplotypes, the human leukocyte antigen region and the SARS-CoV-2 peptidome. By inversion imputation, we traced a reported association at 17q21.31 to a ∼0.9-Mb inversion polymorphism that creates two highly differentiated haplotypes and characterized the potential effects of the inversion in detail. Our data, together with the 5th release of summary statistics from the COVID-19 Host Genetics Initiative including non-Caucasian individuals, also identified a new locus at 19q13.33, including NAPSA, a gene which is expressed primarily in alveolar cells responsible for gas exchange in the lung. Introduction Different studies, including the first genome-wide In the past year, Coronavirus disease 2019 (COVID-19), association analysis (GWAS) by of our Severe COVID-19 caused by the severe acute respiratory syndrome coro- GWAS study group, have shown that genetic predisposi- navirus 2 (SARS-CoV-2), has evolved into a global pan- tion plays a role in COVID-19 susceptibility and severity demic with more than 532 million confirmed cases and (5–7). In particular, we reported significant associations 6.3 million COVID-19-related deaths worldwide (frequen- between genetic variants at loci 3p21.31 (around LZTF1) cies reported by the World Health Organization, June and 9q34.2 (ABO blood group locus) to severe respiratory 11, 2022). The clinical manifestations of COVID-19 are failure and SARS-CoV-2 infection in 1610 severe COVID- variable and range from complete absence of symptoms 19 patients and 2205 population controls of European to severe respiratory failure and death. Severe COVID- ancestry. Since then, these loci have been replicated 19 requires intensive medical care with respiratory sup- in subsequent studies and extended also to non- port and can result in long-term damages detrimental European cohorts (8–10). Eleven additional genome-wide to the individual. The pathogenesis of severe COVID-19 significant loci, associated with SARS-CoV-2 infection is, however, still poorly understood. This condition has or COVID-19 manifestations, have been reported by been associated with clinical risk factors such as old age, various studies, including the Genetics Of Mortality In male sex and comorbidities including diabetes, active Critical Care (GenOMICC) Initiative and, more recently, cancer, hypertension and coronary artery disease (CAD) the largest genetic consortium for COVID-19, the COVID- as well as solid organ transplant or other conditions that 19 Host Genetics initiative (HGI) (6,7). Six of these loci promote an immunosuppressive state (1–4). have been linked to critical illness by COVID-19 and 3950 | Human Molecular Genetics, 2022, Vol. 31, No. 23 include genes that were previously associated with pul- support classes 1–4 with a total of 3255 patients and monary or autoimmune and inf lammatory diseases (7). 12 488 controls, main analysis), and second, a more Since our primary publication, the Severe COVID-19 stringent definition of severe COVID-19 ‘hospitalization GWAS study group dataset has been extended to 3255 with mechanical ventilation’ (classes 2–4 with 1911 severe COVID-19 patients with respiratory failure (5) and critically ill individuals and 12 226 controls, subtype 12 488 controls with unknown COVID-19 status from analysis). Details of per cohort patient numbers are 30 study centers across Italy, Spain, Norway and Ger- shown in Supplementary Material, Table S1E. The many/Austria (after quality control (QC); Supplementary characteristics of patients and controls included in our Material, Table S1A and D). Compared with our previous analyses are shown in Table 1 and Supplementary Mate- study, we increased the number of patients and controls rial, Table S2. After imputation, we carried out a GWAS by approximate factors of 2 and 5.5, respectively (from of 9 223 806 and 9175283 high-quality genetic variants 1610 and 2205). Furthermore, detailed information on [imputation R2 ≥ 0.6 and minor allele frequency (MAF) age, sex and comorbidities as hypertension, diabetes ≥1%] stratified by ancestry (Italy, Spain, Norway and and CAD is available in this dataset. The control cohort Germany/Austria) using a logistic mixed model analysis included predominantly population controls recruited as implemented in SAIGE (12), followed by a fixed-effect at the respective centers as blood or bone marrow inverse variance-weighted meta-analysis using METAL donors with unknown or negative COVID-19 status (13) (low genomic inflation of 1.017; Supplementary (Supplementary Material, Table S1C). With this unique Material, Figs S2 and S3). Genome-wide comparison of resource at hand, we performed an extended genome- the case–control frequencies [conservatively adjusted wide association study (GWAS) and meta-analysis of for age, sex, age∗ age, sex∗ age and top 10 principal severe COVID-19 as well as a meta-analysis with the components (PCs) from principal component analysis COVID-19 HGI release 5 GWAS summary statistics, (PCA) as employed by the COVID-19 HGI; Material and followed by stratified analyses of age, sex and disease Methods], revealed genome-wide significant associations and a detailed analysis of ABO secretor status. Using Y- (P < 5 × 10−8 ; Table 2 and Supplementary Material, Table chromosomal genotype calls from the Global Screening S3, Figs S4 and S5) at three known loci in the main anal- Array (GSA), we additionally performed an imputation ysis: chr3:45848457:C:T (rs35731912) within LZTFL1 at and disease association of chromosome Y haplotypes. the 3p21.31 locus, chr9:133261703:A:G (rs687289) within Compared with our first study, we also performed an ABO at 9q34.2, chr19:10351837:C:T (rs11085725) within even more detailed analysis of the human leukocyte TYK2 at 19p13.2. In addition to the same TYK2 variant, antigen (HLA) which includes classical fine-mapping chr19:4717660:A:G (rs12610495) within DPP9 at 19p13.3 of the HLA region based on local imputation of SNP, was genome-wide significant in the subtype analysis. amino acid and classical allele information, as well as a Additional suggestive loci (P < 10−6 ) from both analyses broad range of other approaches, including a peptidome- include previously reported associations at 17q21.31 wide association study (PepWAS) (11) computational (MAPT), 19p13.3 (DPP9) and 21q22.11 (IFNAR2) (Table 2 prediction of SARS-CoV-2 peptide presentation, HLA and Supplementary Material, Table S3) (7). Of note, the class I supertype association analysis and tests for respective lead risk variants (this study versus COVID- heterozygote advantage, divergent allele advantage and 19 HGI) at reported loci are linked with r2 or D’ > 0.9 molecular mimicry. Finally, by inversion imputation, we across our study cohorts (Supplementary Material, present a functional analysis of the established 17q21.31 Table S4). Although hierarchical mixture model analysis locus (7), which has been described by COVID-19 HGI but with MAMBA (14) (Materials and Methods) showed a not characterized in depth. high posterior probability of replication (PPR > 0.8) for suggestive associations at PCDH7 at 4p15.1, FREM1 at 9p22.3 (main analysis), OLMF4 at 13q21.1 and PTPRM at Results 18p11.23 (subtype analysis), they were not observed as GWAS meta-analyses for severe COVID-19 with significantly associated with COVID-19 in the COVID-19 respiratory failure HGI 5 release dataset, which is why we did not consider Genotyping was carried out using Illumina’s GSA, them further (forest plots including all data are shown followed by genotype QC analysis and TOPMed genotype in Supplementary Material, Fig. S6). Future analyses imputation (Material and Methods, Supplementary will be required to determine the significance of these Material, Figure S1, patient numbers before and after QC associations to COVID-19. are shown in Supplementary Material, Table S1A and C). Next, we performed a replication analysis of the Respiratory failure was defined as respiratory support 13 variants reported to be genome-wide significantly with supplemental oxygen [class 1] or non-invasive associated with COVID-19 by the COVID-19 HGI. Since and invasive ventilation [classes 2 and 3, respectively], data from the Severe COVID-19 group where shared or by extracorporeal membrane oxygenation (ECMO) with the COVID-19 HGI, release 5 of the HGI con- [class 4]).(5) Analogously to the COVID-19 HGI (7), tains 3815 individuals from our previous analysis. We we conducted two GWAS discovery meta-analyses for therefore included only individuals from the Severe two main categories of COVID-19 disease state: First, COVID-19 GWAS group not contained in the COVID- ‘hospitalization with respiratory support’ (respiratory 19 HGI release 5 datasets (Material and Methods and Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3951 Table 1. Overview of patients included in the genome-wide discovery analysis Italy Spain Norway Germany/Austria Total N severe COVID-19 patients 1536 1421 262 241 Respiratory support categories N (%) Supplemental oxygen only (1) 276 (17.97) 897 (63.53) 80.65 109 (45.23) Non-invasive ventilation (2) 933 (60.74) 122 (8.62) 3.23 21 (8.71) Ventilator (3) 320 (20.83) 381 (26.91) 16.13 86 (35.68) ECMO (4) 7 (0.46) 16 (1.13) 0 25 (10.37) Median age (IQR)—years (1) 75 (21) 69 (21) 59 (20) 63 (23) Median age (IQR)—years (2) 66 (23) 72 (17) 68.5 (10) 68 (20) Median age (IQR)—years (3) 63 (14) 65 (15) 66 (19) 64 (18) Median age (IQR)—years (4) 49 (12) 57 (10) / 56 (12) Main analysis N severe COVID-19 patients 1536 1421 262 241 Female sex—% 32.81 35.88 33.87 26.97 Median age (IQR)—year 67 (22) 68 (18) 59.5 (21) 63 (19) Hypertension—% affected (% missing) 40.55 (21.16) 49.6 (1.91) 41.3 (25.81) 53.09 (32.78) CAD—% affected (% missing) 17.09 (21.16) 10.16 (1.98) 25.81 (0) 15.23 (37.34) Diabetes—% affected (% missing) 14.86 (21.16) 24.55 (1.91) 14.52 (0) 19.63 (32.37) Subtype analysis N severe COVID-19 patients 1260 519 132 Female sex—% 29.68 27.55 / 22.73 Hypertension % affected (% missing) 38.26 (23.25) 52.19 (3.28) / 62.96 (38.64) CAD—% affected (% missing) 16.03 (23.25) 12.55 (3.28) / 18.75 (39.39) Diabetes—% affected (% missing) 14.58 (23.25) 25.3 (3.28) / 19.75 (38.64) Overview of patients included in our first analysis (3260 patients) and second analysis (1911 patients). Individuals of the Italian, Spanish, Norwegian and German/Austrian cohorts were recruited at 5, 7, 8 and 10 different hospitals/centers, respectively. Shown are respiratory support status groups 1–4, age and median age across all individuals as well as within each respiratory support group, percentage of females within each cohort, as well as percentage of individuals affected by known comorbidities of COVID-19. Commonly reported comorbidities in COVID-19 are shown, hypertension, CAD and diabetes. Characteristics of control individuals are shown in Supplementary Material, Table S2. Table 2. Candidate variants from the main and subtype analysis snpid rsid A1 A2 FRQ OR OR_95L OR_95U PPR P gene Main analysis chr3:45848457:C:T rs35731912 T C 0.09 1.784 1.723 1.847 1.000 2.32E−21 LZTFL1 chr19:10351837:C:T rs11085725 T C 0.273 1.269 1.254 1.283 1.000 1.02E−10 TYK2 chr9:133261703:A:G rs687289 A G 0.358 1.240 1.219 1.262 1.000 4.49E−10 ABO chr19:4717660:A:G rs12610495 G A 0.27 1.216 1.188 1.244 0.990 9.91E−08 DPP9 chr9:14754866:A:G rs7023573 A G 0.285 1.207 1.185 1.231 0.954 5.21E−07 FREM1 chr17:45933112:G:A rs8065800 G A 0.328 1.194 1.145 1.247 0.925 8.90E−07 MAPT chr4:30946048:T:C rs12512667 C T 0.444 0.850 0.828 0.873 0.922 8.95E−07 PCDH7 Subtype analysis chr19:10351837:C:T rs11085725 T C 0.272 1.309 1.294 1.325 0.999 3.24E−09 TYK2 chr19:4717660:A:G rs12610495 G A 0.267 1.286 1.257 1.316 0.993 2.87E−08 DPP9 chr13:55031860:T:C rs111671068 C T 0.012 2.802 2.794 2.810 0.806 1.43E−07 OLFM4 chr9:133271182:T:C rs550057 T C 0.266 1.262 1.238 1.286 0.939 4.56E−07 ABO chr18:7897309:A:C rs17565758 C A 0.078 1.510 1.470 1.550 0.938 4.84E−07 PTPRM chr21:33247902:C:T rs2834161 C T 0.371 1.250 1.175 1.331 0.923 6.10E−07 IFNAR2 We show only variants that have a PPR > 0.8 in the MAMBA analysis. A more detailed overview of all variants is shown in Supplementary Material, Table S3. snpid: id of the marker as chr:pos:alleles, genome build hg38; A1: minor allele from TopMed imputation; A2: major allele from TopMed imputation; OR/OR_95L/OR_95U: OR and 95% CI of A1; PPR: posterior probability of replication; P: P value of association. Supplementary Methods) for this analysis. 8 of the 13 analyses as well as 9 163 456 and 9 309 373 high quality variants reported as genome-wide significant by the variants (imputation R2 ≥ 0.6 in our cohorts and overall COVID-19 HGI (7) replicated at least a nominal P-value MAF ≥ 1%; Supplementary Material, Figs S7 and S8). of 0.05 (located near LZTFL1, ABO, TYK2, DDP9, IFNAR2, This analysis prioritized rs1819040 at the 17q21.31 MAPT/KANSL1, OAS1; Supplementary Material, Table S5). locus within the KANSL1 gene as the most strongly Subsequently, we performed a fixed-effect inverse associated variant in this region (P = 3.27 × 10−11 for variance-weighted meta-analysis using METAL (13) rs1819040; OR = 0.88 for minor allele T; 95%CI = 0.84–0.92) across the main (hospitalized with respiratory support) over rs8065800 from the main analysis (Supplementary analysis and COVID-19 HGI B2 statistics as well as Material, Tables S3, S4 and S6). The meta-analysis the subtype analysis (critically ill: hospitalized with of our subtype analysis with the COVID-19 HGI A2 mechanical ventilation) and COVID-19 A2 statistics, summary statistics revealed an additional genome-wide resulting in a total of 1 320 760 (14 467 and cases significant locus not previously associated with severe and 1 306 293 controls) and 725 601 (6526 cases and COVID-19, chr19:50379362:T:C (rs1405655, P = 3.25 × 10−8 ; 719 075 controls) analysed individuals in the respective OR = 1.09; 95%CI = 1.06–1.13) located near the NAPSA 3952 | Human Molecular Genetics, 2022, Vol. 31, No. 23 Figure 1. Forest plot of candidates from the in-depth stratified analysis. The plots show the variants chr3:45823240, chr3:45848457:C:T, chr17:46142465:T:A which were significantly associated with age (interaction P-value [P_FDR(I) < 0.05]) when comparing age groups ≤60 and >60. We additionally show variants chr19:4717660:A:G and chr21:33242905:T:C with insignificant, though strong trends for association with age. ORs and their respective 95% confidence intervals (CIs) are visualized for each of our four cohorts separately. The size of the dots indicates the size of the respective cohort (N). The OR value is displayed in addition as a numerical value. Only cohorts in which N > 50 in both cases and controls are shown. The headers are built as follows: gene variant id (chr:pos(hg38):allele)—rs-id—effect allele ∗/∗∗ . ∗ indicates a variant that was observed as associated from data in this study; ∗∗ indicates a variant that was observed as associated in the release 5 of the COVID-19 HGI. gene at 19q13.33 (regional association and forest plots such as hypertension, CAD and diabetes in cases are shown in Supplementary Material, Fig. S9, Table only and performed sex-stratified analysis within age S6). Bayesian fine-mapping analysis of this locus with groups and with respect to disease severity groups. We FINEMAP (15) (Material and Methods) identified a total performed interaction analysis of the SNP with sex and of 15 (log10 (Bayes factor) = 5.95) variants that belong to age (age group ≤ 60 and >60), respectively, to investigate the 95% most likely to be causal (Fig. 1, Supplementary sex-specific and age-specific effects, i.e. to determine Material, Table S7), indicating that a small number of whether differences in effects within these groups SNPs could be considered causal in subsequent analyses. were statistically significant (Materials and Methods and Supplementary Methods). Results of these analyses are In-depth stratified analysis of lead variants shown in Figure 1, Supplementary Material, Tables S8 To estimate differences in genetic effects across different and S9, Figures S10 and S11. We confirmed our previously age groups and different biological sex categories, we described associations with age (Fig. 1) and with disease performed an in-depth stratified analysis for selected severity at 3p21.31 for chr3:45848457:C:T (severity: variants. These included genome-wide and suggestively P = 9.73 × 10−7 , OR = 1.65; 95%CI = 1.35–2.03), which is significant variants from this study (LZTFL1, ABO, TYK2, also discussed in detail in a study by Nakanishi et al. DPP9; MAPT, IFNAR2; 7 lead variants in total from (16). Differential association with age was also observed the main and the subtype analysis). We additionally for chr17:46142465:T:A (rs1819040, MAPT) (Fig. 1). Strong included all of the 13 variants from the COVID-19 but non-significant trends for an association were also HGI release 5 analysis (7) and the novel association, observed for chr19:4717660:A:G (rs12610495; DPP9) at chr19:50379362:T:C, at the NAPSA locus. A detailed the 19p13.3 locus, and chr21:33242905:T:C (rs13050728; overview of these loci is shown in Supplementary IFNAR2) which was significantly associated with severe Material, Table S8. We additionally investigated the COVID-19 only in younger individuals before correction association of these variants to known comorbidities for multiple testing (Fig. 1). We analysed these 5 variants Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3953 even more in-depth within the age groups of 41–50, 51– Functional analysis of 17q21.31 and 19q13.33 60 and 61–70 as well as 41–60 and 61–80 (Supplementary using publicly available datasets Material, Fig. S10C) and observed a tendency for stronger We next performed several follow-up analyses to better effects in the younger population at the four loci. understand possible functional implications of the Interestingly, the German population showed opposite associations at the 17q21.31 inversion and at the novel effect sizes than the other cohorts in the broad age group 19q13.33 locus, including a phenome-wide association analysis for chr17:46142465:T:A (see also below). study (PheWas) and exploratory gene expression We did not observe any statistically significant asso- analysis. ciations with age, sex, severity or comorbidities for the We queried variants in high LD (r2 > 0.9) with chr19:50- remaining analysed lead variants. Additional age and 379362:T:C (rs1405655) or the 17q21.31 inversion, using a sex stratified forest plots of this analysis are shown in wide range of phenotypes from the NHGRI GWAS Cat- Supplementary Material, Figure S11A and B. alog (Material and Methods) (19). While no known phe- As the genome-wide significant association near notypes were found to be linked to chr19:50379362:T:C LTZF1, TYK2, DDP9 and ABO have been followed up in (rs1405655) or its proxy variants, 161 GWAS associations detail before, we next focused on the novel association were identified for variants in high LD (r2 > 0.9) with at the 19q13.33 (NAPSA) locus from the meta-analysis the 17q21.31 inversion, illustrating its multiple effects with COVID-19 HGI release 5 summary statistics and (Supplementary Material, Table S11). These associations performed an in-depth analysis of the 17q21.31 (MAPT, included several traits potentially related to COVID-19 KANSL1) locus associated with age. The 17q21.31 locus pathology, such as blood and immune cell composition has been previously reported to be linked to a common or lung function (Fig. 2C). inversion polymorphism of two highly divergent haplo- Variants within the credible set from Bayesian fine types H1 and H2 (17). Indeed, Bayesian fine-mapping mapping at 19q13.33 overlap with several genes, includ- (15) at this locus showed that the 95% credible set ing for instance NAPSA, NR1H2 and KCNC3, while the includes 1530 variants among these chr17:45933112:G:A 17q21.31 inversion spans multiple genes including MAPT, (rs8065800) from our initial discovery analysis and KANSL1, FMNL1 and CRHR1 (Supplementary Material, chr17:46142465:T:A (rs1819040) from the COVID-19 HGI Figs S4, S5 and S9 and Fig. 2). To identify possible analysis with certainty <0.3%, indicating, based on target genes relevant in COVID-19 disease pathology, we the overlap of LD boundaries of highly associated SNP performed an exploratory gene expression analysis using variants with the inversion boundaries (Fig. 2A), that the several publicly available datasets to: (i) examine the individual SNP associations may be proxy variants for direct effect of both loci on gene expression by analysing the associated inversion polymorphism. colocalization of lead SNPs with expression and splicing We therefore imputed the inversion haplotypes quantitative trait loci (eQTL and sQTL), (ii) identify in H1 and H2 for our cohorts by genotype imputation which tissues or cell types our candidate genes are with IMPUTE2 (18), employing as reference 109 indi- expressed by analysing their RNA expression at bulk and viduals from the 1000 Genomes Project for which single-cell levels and (iii) infer the possible contribution 17q21.31 inversion genotypes were obtained exper- of these genes to COVID-19 pathology by looking at imentally by FISH and droplet digital polymerase- their expression patterns in (a) monocytes exposed to chain-reaction (PCR; Material and Methods). LD between different viral and non-viral immune stimulators; (b) the chr17:46142465:T:A (rs1819040) variant, which organoids infected with COVID-19 and (c) single-cell was prioritized over chr17:45933112:G:A (rs8065800) in RNA-seq of lung and other tissues from patients who the meta-analysis with the COVID-19 HGI summary died after experiencing a SARS-CoV-2 severe disease statistics, and the inversion in our cohorts is near perfect (Material and Methods). Results of these analyses are (r2 = 0.98, D = 0.99). Genome-wide significant association displayed in Figure 3. with severe respiratory COVID-19 for the inversion The potential functional role of the 17q21.31 inversion was confirmed using logistic regression followed by is supported by the fact that the inversion locus meta-analysis across this study’s panels and summary (chr17:45883776 ± 75 kb) strongly colocalizes (regional statistics derived for the from the COVID-19 HGI data probability > 0.9) with eQTLs and sQTLs (i.e. displaying (Materials and Methods) (meta-analysis main analysis the strongest association with the target), for 36 (eQTL) and COVID-19 HGI release 5 B2: P = 7.61 × 10−10 , OR = 0.89; and 15 (sQTL) genes, respectively (Fig. 3A, Supplementary 95%CI = 0.84–0.92; meta-analysis subtype analysis and Material, Table S12, Fig. S12). Expression patterns of COVID-19 HGI release 5 A2: P = 1.5 × 10−4 , OR = 0.90; protein-coding genes with eQTLs associated to the 95%CI = 0.85–0.95; Fig. 2B, Supplementary Material, Table inversion are shown in Supplementary Material, Figure S10). Using stratified analysis, we also show an age effect S13, with the best candidates to play a role in the for the inversion (Supplementary Material, Table S7, effects of the inversion on COVID-19, including MAPT, Fig. S10). The opposite effects direction observed for KANSL1, FMNL1 and CRHR1, being summarized in the German cohort for linking SNP chr17:46142465:T:A Figure 3. Many of these genes are highly expressed in was shifted for individuals aged 40–61 when analysing neural tissues and testis (Supplementary Material, Fig. the inversion. We did not observe any association of the S13). However, several of them are also expressed in inversion with disease severity. major immune cell types in COVID-19 relevant tissues 3954 | Human Molecular Genetics, 2022, Vol. 31, No. 23 Figure 2. Association of the 17q21.31 locus with severe COVID-19 with respiratory failure. (A) Regional association plot showing the variant most strongly associated with severe COVID-19 (rs1819040, purple diamond) a ∼ 0.9 Mb inversion polymorphism at 17q21.21 (56) (white line with blue rectangles representing the variable segmental duplication (SD) blocks at the breakpoints), and the large credible set obtained by statistical fine-mapping including 2178 SNPs in high LD (median [LD] = 0.97) with the inversion (Supplementary Material, Table S7). Pairwise LD values (r2 ) with lead variant rs1819040 were calculated from merged Italian, Spanish, German/Austrian and Norwegian GWAS main anlysis datasets. The dotted line indicates the genome- wide significance threshold (P = 5 × 10−8 ). Below, organization of the 17q21.31 inversion genomic region, with the extended haplotypes associated with each orientation (H1 and H2) shown as red and blue arrows, respectively, and breakpoint SDs as dark rectangles. Protein-coding genes for which the inversion is a lead eQTL in at least one GTEx tissue are shown as pointed rectangles indicating the direction of transcription. (B) Forest plot and extended meta-analysis of our first discovery analysis and the COVID-19 HGI release 5 analysis B2 dataset (Material and Methods) of the association between severe COVID-19 and the 17q21.31 inversion based on the presence relative to the absence of the inversion haplotype H2. We visualized the ORs and their 95% confidence invervals (CIs) across all analysed cohorts of the main analysis and the COVID-19 HGI release 5 analysis B2 data. In this analysis, the overlap between the main analysis cohort and the COVID-19 HGI data was excluded from the main analysis cohort. The OR value is displayed in addition as a numerical value. The size of the dots indicates the size of the respective cohort (N). (C) Phenome-wide association study (PheWAS) results for the 17q21.31 inversion allele H2 showing only potentially COVID-19-related phenotypes from the GWAS Catalog (P = 10−7 ) grouped by disease categories using different colors. The effect direction of known SNP-trait associations from the corresponding GWAS is shown using triangles pointing upward (increase) and downward (decrease), whereas dots represent unknown effect direction. Phenotypes shown were selected according to previously reported COVID-19 links with lung damage, blood cell alterations and exacerbated immune response, as well as some potential co-morbidities. The whole list of phenotypic associations is included in Supplementary Material, Table S11. (Fig. 3B, Supplementary Material, Fig. S13A/B), such Table S13). For example, KANSL1, that could have anti- as KANSL1, which is expressed in lung tissue-resident inflammatory effects, shows higher expression in H2 alveolar macrophages, or FMNL1 (Fig. 3B, Supplementary carrier monocytes stimulated with Influenza A virus Material, Fig. S13B). These two genes especially show sig- infection-like conditions which appears to be related nificantly higher expression in different lung cell types in to a significant increase of coding and non-coding acute COVID-19 patients who died with the disease as isoforms (Supplementary Material, Fig. S15). Similarly, compared with healthy controls (20) (Fig. 3C, Supple- in SARS-CoV-2 infected brain organoids, the expression mentary Material, Fig. S13C). Moreover, RNA-seq data of MAPT was significantly downregulated in premature of monocytes under different bacterial and viral stimuli and mature neuronal cells (Supplementary Material, Fig. that activate toll-like receptor pathways, including bacte- S16, Table S14). rial lipopolysaccharide, Pam3 CSK4 , R848 and inf luenza A In the case of the 19q13.33 locus, expression of candi- virus, show expression changes in several genes affected date genes shows high tissue specificity, with NAPSA by the inversion (21) (Supplementary Material, Fig. S14, mRNA being specific to lung and lung parenchyma Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3955 Figure 3. Expression analysis of the most plausible candidate genes associated with the 17q.21.31 and 19q.13.33 loci in organ tissues and COVID-19 relevant cell types. (A) GTEx tissue-specific expression QTL (eQTL, upper panel) and splicing QTL (sQTL, middle panel) effects of the 17q21.31 and 19q.13.33 loci on selected candidate genes as well as expression of these genes in GTEx (20) tissues (lower panel). The direction of the normalized eQTL and sQTL effect size (NES) of the lead SNP rs1405655 and the inversion tagger rs62055540 in perfect LD with the inversion is represented by color intensities, and statistical significance by dot size. Black rectangles indicate genes for which the expression colocalizes (regional probability > 0.9) with GWAS loci in a given human tissue from the GTEx dataset. Heatmap displays gene-wise centered median by tissue expression values (represented by color intensities), showing in which tissues candidate genes are mostly enriched. (B) Expression levels of candidate genes in scRNA-seq datasets from healthy upper airways (nasal, bronchi) and lung (parenchyma) cells (47) and adult human brain cells from recently deceased, non-diseased donors (48). Figure displays log-normalized mean expression (represented by color) and fraction of cells expressing those genes (indicated by dot size). Processed and cell-type-annotated gene expression levels from studies were retrieved from COVID-19 Cell Atlas (49). (C) The figure shows differential expression of candidate genes in lung cells of COVID-19 patients compared with healthy controls. Log2 fold change (log2FC) values are presented as color gradient. Nominal P-values in −log10 scale are shown proportionally to dot size. Black-bordered circles indicate significantly differentially expressed genes after FDR correction. Results were obtained from pseudo-bulk differential expression analysis by Delorey et al. (21). More detailed figures are shown in Supplementary Material, Figures S11, S12 and S15. and KCNC3 being highly expressed in brain and thyroid appears to be clearly affected by the rs1405655 lead SNP tissue, while NR1H2 is more broadly expressed among which is also displayed by strong colocalization [regional human tissues, including many immune cell types probability > 0.9] of the loci with eQTLs in multiple (Fig. 2B, Supplementary Material, Fig. S12). Of those tissues (Fig. 3A). Single SNP Mendelian randomization candidates, expression of KCNC3 and especially NAPSA analysis using cis-eQTL data and assessment of effect 3956 | Human Molecular Genetics, 2022, Vol. 31, No. 23 heterogeneity identified higher expression of NAPSA Y chromosome describe so-called Y-chromosome hap- as a potentially causal protective factor for severe logroups with letters A–Z (defined by the Y Chromosome COVID-19 (Beta = −0.39; P = 1.26 × 10−7 ) (Supplementary Consortium) (23) and follow a pattern of ancestral popu- Material, Table S15). Notably, NAPSA shows significantly lation migrations in Europe and on a global scale. increased expression in type 1 alveolar cells of COVID-19 Results of the Y-chromosome haplogroup analysis are patients as compared with healthy controls (20) (Fig. 3, shown in Supplementary Material, Table S20. We did not Supplementary Material,Fig. S13). On the other hand, observe any statistically significant or consistent results the NR1H2 gene is significantly down-regulated in some for Y chromosomal haplotype association with COVID- parenchymal (basal, ciliated, club) and endothelial 19 across the Spanish, Italian and German populations in (pericyte) lung cells and is up-regulated in monocytes of our main or subtype analysis after correction of P-values COVID-19 patients when compared with healthy controls for multiple testing using Bonferroni. We found that (Fig. 3C, Supplementary Material, Fig. S13) (20). cohort-specific suggestive (P < 0.05) associations within the Italian population for the R1b haplogroup were par- tially driven by single genotyping batches with diverging Association analysis of specific candidate haplogroup frequencies (i.e. high coefficient of variation genomic regions: ABO locus, HLA locus and between control batches). COVID-19-related mortality Y-chromosome haplogroups showed a strong trend for association with haplogroup The ABO locus has one of the most significant associa- R1b1a2a1 (U106) (P = 9.49 × 10−3 , OR = 2.8, 95%CI = 1.29– tions with COVID-19 infection, as highlighted in several 6.2) in both the Spanish and Italian populations. This publications (5,7). In our initial publication, we showed association remained after adjusting for the comorbidi- that the genetic association at the ABO locus could be ties hypertension, CAD and diabetes. COVID-19-related explained by differences in the actual ABO blood type. death remains, however, a challenging endpoint influ- ABO secretor status, i.e. secretion of ABO antigens into enced by many factors. body f luids, is determined by genetic variation in the FUT2 gene. Recent studies suggest that A secretors have a higher risk of COVID-19 susceptibility (22). Here, we Discussion sought to replicate this finding (Material and Methods) We here present a large collaborative COVID-19 genetics 81% of all individuals were secretors irrespective of blood study of different centers from Italy, Spain, Norway and group on average across all control cohorts. In our study, Germany/Austria. With our clearly defined phenotype though not statistically significant in an interaction anal- of severe respiratory COVID-19, a centralized genotyp- ysis of blood group and secretion status, a trend was ing and rigorous QC, we have generated a valuable observed for increased risk of COVID-19 infection for A resource for further COVID-19-related genetic studies. By secretors when comparing COVID-19 cases to the general conservatively including only severely affected COVID- population (A secretors: OR = 1.32; 95%CI = 1.20–1.46; A 19 patients in our analysis, we aimed to balance out non-secretors: OR = 1.08; 95%CI = 0.91–1.28) (Supplemen- a potential bias from selection of controls from the tary Material, Table S16). B secretion was not significantly general population, which is however general practice in associated with COVID-19 susceptibility and both A/B GWA analyses. Our study replicates known associations secretion were not significantly associated with disease with COVID-19 from our previous and other studies severity. and suggests associations at four novel loci, PCDH7 at The HLA fine-mapping approach yielded no associ- 4p15.1; FREM1 at 9p22.3; OLMF4 at 13q21.1 and PTPRM ation at the genome-wide or nominal (P < 10−5 ) signif- at 18p11.23. These associations do not replicate in the icance threshold, neither in the overall meta-analysis COVID-19 HGI release 5 statistics despite their high across the four cohorts nor within the separate cohorts level of replicability across all our different cohorts. (Supplementary Material, Table S17, Fig. S17). Further- Future studies in independent cohorts will be required more, we found no significant association for any HLA- to confirm their association to COVID-19 disease. presented viral peptide in a so-called PepWAS approach Genome-wide meta-analysis of COVID19 HGI release (Supplementary Material, Table S18, Figs S18 and S19), 5 summary statistics with our data revealed a so-far where associations between HLA-presented peptides and unreported association at the 19q13.33 locus, near the disease are unraveled by integrating similarities and dif- gene NAPSA. Our in-depth stratified analysis showed sig- ferences in peptide binding among HLA alleles across nificant age-associations for LZTFL1 at the 3p21.31 locus patients nor robust statistical associations with any of and the inversion at 17q21.31. Effects of age in genetic the other tested HLA parameters (Supplementary Mate- variation are frequently observed in genetic studies; how- rial, Table S19). ever, reasons for this are not clear. Possible models are With male sex identified as a risk factor for severe discussed in detail in Jiang et al. (24). COVID-19 and COVID-19-related death (3), we explored Our functional analysis focused on two loci of interest possible associations between genetic variants on the Y associated with severe COVID-19, the previously known chromosome and the risk of developing severe COVID-19 17q21.31 inversion and the 19q13.33 locus, including the and COVID-19 mortality in males (3). Variations on the NAPSA gene. For the novel association at the 19q13.33 Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3957 locus, additional analyses provided first hints for a func- function of many of the affected genes is not well tional involvement in COVID-19 through its regulation known, the inversion acts as an eQTL and sQTL of of the NAPSA gene. NAPSA encodes for a protease highly several interesting candidate genes for severe COVID- expressed in Type 1 (AT1) and Type 2 (AT2) alveolar cells, 19. In particular, there are several genes potentially where two cell types required for the gas exchange at the associated with immune function and immune response. lung surface and the secretion of surfactant proteins as For example, KANSL1, involved in histone acetylation, well as immunomodulatory factors (AT2) (25). Comple- is broadly expressed in many types of immune cells mentary findings were observed in another recent study in upper airways and lung tissue (Fig. 3) and has been dissecting the lung transcriptome of COVID-19 infected proposed to play a role in the macrophage transition to patients in which NAPSA expression is increased in AT1 an anti-inflammatory phenotype in mice (33). Here, we cells. This study also linked NAPSA to the marker gene have found that the expression decrease in infection- expression signature of ‘damage-associated transient like stimulated monocytes is partially compensated in progenitors’, an intermediate cell state between AT1 homozygotes for the H2 inversion haplotype. CHRH1, and AT2 cells promoted by inf lammation, characterized which is associated to higher expression in the H2 by a failure of AT2 cells to differentiate to AT1 cells haplotype in several tissues (Fig. 3), encodes a receptor (26). Thus, given that the NAPSA protein is involved in that binds to corticotropin-releasing hormones, which lung surfactant production, which is dysregulated in are major regulators of the hypothalamic–pituitary– COVID-19 (27), the fact that the COVID-19 risk allele adrenal axis, and regulates immune and inflammatory is associated with decreased NAPSA expression, while responses (34). Finally, FMNL1, which is also located increased expression of NAPSA is a protective factor for in the inversion locus, shows high expression levels in severe COVID-19, suggests a potential role of NAPSA in macrophages, dendritic cells and B and T lymphocytes susceptibility for severe COVID-19. These findings are in different COVID-19-related tissues (Fig. 3) and it is in line with a recent report in which NAPSA has been involved in cell motility and T cell trafficking (35). In identified as a candidate gene for COVID-19 severity addition, many of the genes in the 17q21.31 inversion through a cross methylome omnibus test combined with regions show a predominant expression in brain tissues, S-PrediXcan analyses in blood tissue and integrative which could also play an important role in COVID-19. multiomics (28), as well as with another recent report The clearest example is MAPT, which is downregulated in which summary-based Mendelian randomization in SARS-CoV-2 infected neuronal cells (Supplementary analysis showed an inverse correlation between the Material, Fig. S16) and lung-related cells and tissues. The expression of NAPSA in blood and COVID-19 severity H2 haplotype is linked to increased expression of MAPT in (29). the lung and its MAPT exon 3 in brain tissues (36), which Though we cannot prove causality of the 17q21.31 could compensate for the downregulation during viral inversion polymorphism in COVID-19 disease, it is infection and have a protective effect against COVID- linked to many traits potentially relevant to COVID-19 19. However, despite the potential implication of these outcome. For instance, the inverted haplotype H2 was and other genes, it is not possible to single out just one previously associated with higher number of red blood as the most likely candidate. In fact, inversions are well- cells and hemoglobin levels, whereas each haplotype known for keeping together a combination of alleles from correlates with different proportions of lymphocytes different genes that generate complex phenotypic traits and granulocytes, which could potentially modulate the in different organisms (37,38). immune response during SARS-CoV-2 infection (30). In Our hypothesis-driven analysis of associations in the addition, the H2 protective allele is also associated with HLA, as well as COVID-19-specific PepWAS analyses, decreased lung function and increased risk of chronic yielded no significant results, indicating no major role obstructive pulmonary disease but protects against the for HLA variability in mediating the severity of COVID-19 development of pulmonary fibrosis (31), and is associated in our cohorts. These results are in line with a recent, with higher ventilatory response to corticosteroids in and so-far the largest, HLA analysis from Shachar et al. individuals with asthma (32), showing potential trade- (39). Within our analysis of Y-chromosome haplogroups, offs and shared pathways that may be important in none of the results remain significant after correction lung health. This variant has been proposed to be under for multiple testing, though single haplogroups showed positive selection in Europeans through its effect on trends for association (Supplementary Text) and larger fertility (17). Our results point to a role of this polymor- study samples are necessary to obtain reliable conclu- phism in immunity and virus infection defence as well. sions. For individual haplogroups, we observed different Interestingly, inversion effects were found to be stronger frequencies between batches that may also arise from in the younger age group in both severity classes, which different versions of the same genotyping platform. could explain the weaker association in the HGI more In general, the analysis of Y-chromosomal disease- severe A2 phenotype due likely to a larger proportion of relevant SNPs is mainly neglected in GWAS, hence the older individuals (7). The inversion probably affects the curation of Y-chromosomal SNPs on genotyping arrays COVID-19 disease course through its large effects on gene is potentially also a confounding factor. Therefore, to expression shown by us and others (7). Although the gain more knowledge regarding the potential role of the 3958 | Human Molecular Genetics, 2022, Vol. 31, No. 23 Y-chromosome haplogroups in disease in general, this research. For some severely ill patients, an exemption should be investigated in more depth. from informed consent was obtained from a local ethics In summary, our findings add to the number of committee or according to local regulations to allow the genome-wide significant hits for COVID-19—totalling use of completely anonymized surplus material from now 14 independent loci—and provide new insights diagnostic venipuncture. Centers from which samples to the molecular basis of COVID-19 severity that were obtained are listed together with their ethics could potentially trigger subsequent and more targeted approval reference numbers from each ethic committee experiments to develop therapies for severe COVID-19. in Supplementary Material, Table S1b. Materials and Methods Sample processing and genotype calling Study participants and recruitment DNA extraction was either performed at the Institute of Clinical Molecular Biology (IKMB, Christian-Albrechts- We recruited 5228 patients with mild to severe COVID-19, University of Kiel, Germany) or the respective study cen- which was defined as hospitalization only (mild) or with ters (Supplementary Material, Table S1a, Supplementary respiratory failure (severe) with a confirmed SARS-CoV-2 Methods) from whole-blood or buffy coat samples and viral RNA PCR test from nasopharyngeal swabs or other for a very small subset also saliva. Genotype calling relevant biologic f luids, cross sectionally, from intensive was performed at the IKMB for all samples with the care units and general wards at different hospitals from Illumina GenomeStudio Version 2.0 software using Italy (4 centers, N = 1857), Spain (6 centers, N = 2795), cluster definition files GSAMD24v2-0_20024620_A1- Norway (7 centers, N = 127) and Germany/Austria (8 762Samples-LifeBrain (GSA Version 2.0; 712 189 SNPs), German, 1 Austrian center, N = 449). For comparison, GSAMD-24v3–0-EA_200034606_A1 (GSA Version 3.0; we included 13 705 control participants from Italy (4 730 059 SNPs) and GSAMD-24v1-0-A_4349HNR_Samples centers, N = 5247), Spain (3 centers, N = 4552), Norway (GSA Version 1.0; 700 078 SNPs). After genotyping, a (1 center, N = 288) and Germany (1 center, N = 3582). total of 4067 German/Austrian (449 cases/3618 con- Details on the centers and origin of the control panels trols), 7347 Spanish (2795 cases/4552 controls), 415 are shown in Supplementary Material, Table S1a and Norwegian (127 cases/288 controls) and 7114 Italian b. Though all patient samples that were sent to our (1857 cases/5247 controls) samples were available with study center were processed, only the severe COVID-19 non-missing core-phenotype (COVID-19 case–control individuals were analysed in this study (Supplementary status) information. For individuals with missing sex Material, Table S1e). Respiratory failure was defined information, sex was inferred from the genotypic sex if in the simplest possible manner to ensure feasibility: possible. the use of oxygen supplementation or mechanical ventilation, with severity graded according to the maximum respiratory support received at any point SNP and sample QC and PCA during hospitalization (1: supplemental oxygen therapy Based on initial genotype data, we removed samples only, 2: noninvasive ventilatory support, 3: invasive with <90% call rate using PLINK (40,41). We additionally ventilatory support or 4: extracorporeal membrane removed individuals with non-matching genotypic and oxygenation) (5). phenotypic sex. After genotype calling, a QC procedure was carried out for the Spanish, Italian, Norwegian Recruiting centers and ethics committee and German/Austrian case–controls GWAS datasets, approval IDs respectively. Variants that had >2% missing data, an The project protocol involved the rapid recruitment of MAF < 0.1% in disease sets or in controls, different patient-participants and no additional project-related missing genotype rates in affected and unaffected indi- procedures (we primarily used material from clinically viduals (PFisher < 10−5 ) or deviated from Hardy–Weinberg indicated venipunctures) and afforded anonymity, owing equilibrium (with a false discovery rate (FDR) threshold to the minimal dataset collected. Differences in recruit- of 10−5 in controls) (i) across the entire collection with ment and consent procedures among the centers arose at most one batch being removed or (ii) falling below because some centers integrated the project into larger in two single batches, were excluded. Samples that had COVID-19 biobanking efforts, whereas other centers did overall increased/decreased heterozygosity rates (i.e. ±5 not, and because there were differences in how local SD away from the sample mean) were removed. For ethics committees provided guidance on the handling robust duplicate/relatedness testing (IBS/IBD estimation) of anonymization or deidentification of data as well and population structure analysis, we used a linkage as consent procedures. Written informed consent was disequilibrium (LD)-pruned subset of SNPs on the basis obtained, sometimes in a delayed fashion, from the of a set of independent (MAF ≥ 5%) SNPs excluding X- study patients at each center when possible. In some and Y-chromosomes, SNPs in LD (leaving no pairs with instances, informed consent was provided verbally or by r2 > 0.2) and 11 high-LD regions as described by Price the next of kin, depending on local ethics committee et al. (42). Pairwise percentage IBD values were computed regulations and special policies issued for COVID-19 using PLINK. By definition, Z0: P(IBD = 0), Z1: P(IBD = 1), Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3959 Z2: P(IBD = 2), Z0 + Z1 + Z2 = 1 and PI_HAT: P(IBD = 2) + 0.5 GWAS and meta-analysis of our main and ∗ subtype cohorts P(IBD = 1) (proportion IBD). One individual (the one showing greater missingness) from each pair with Using genome-wide SNP information, we tested for phe- PI_HAT > 0.1875 was removed. A value of 0.1875 (pro- notypic associations with allele dosage data separately portion IBD) corresponds to a theoretical relationship of for the Italian, Spanish, German/Austrian and Norwegian halfway between the expected IBD for third- and second- case–control data. Including age, biological sex, age∗ age, degree relatives. To identify ancestry outliers, i.e. subjects biological sex∗ age and the first 10 PCs from PCA as covari- of non-European ancestry, we performed PCA for the ates, we performed a logistic association analysis assum- remaining QCed cases and controls including reference ing additive effects with SAIGE for chromosomes 1–22 samples from the 1000 Genomes Phase 3 reference an X (12). For chromosome X association analysis, hap- panel (43). We used the PCA method, as implemented lotypic allele calls outside the pseudo-autosomal regions in FlashPCA (44) on an LD-pruned subset of SNPs (see (PARs) in males are converted to homozygous calls by text above). Ancestry outliers not matching European doubling the haplotypic allele (assuming inactivation of populations were removed (Supplementary Material, large parts of 1 of the 2 female X chromosomes (46) Fig. S1A–F). After QC, PCA revealed no non-European and sex is used as a covariate for association testing of ancestry outliers (Supplementary Material, Fig. S1E–H) the non-PAR regions on chromosome X. For each of the when performing PCA. analyses, sample numbers are given in Supplementary Material, Table S1e. SNP genotype imputation An inverse-variance weighted fixed-effects meta- analysis was conducted with the meta-analysis tool The QCed Italian, Spanish, Norwegian and German METAL (13) on the cohort of the main analysis including GWAS datasets comprised 1563 Italian COVID-19 cases, 3255 cases and 12 488 population controls of unknown 4759 Italian controls, 2174 Spanish COVID-19 cases, 4406 COVID-19 status from Italy, Spain, Norway and Ger- Spanish controls, 81 Norwegian cases, 283 Norwegian many/Austria and the cohort from the subtype analysis controls, 336 German COVID-19 cases and 3303 German including 1911 critically ill cases and 12 226 population controls, and contained 567 131 (Italy), 564 856 (Spain), controls (Supplementary Material, Table S1e). Only 525 836 (Norway) and 476 562 (Germany/Austria) variants variants that were common to at least 2 datasets with after QC and filtering of SNPs with alleles AT or respective post-imputation R2 ≥ 0.6 and that had an CG (the latter often leading to strand issues during overall MAF of ≥1% were considered in the analysis. For imputation). Genotype imputation was conducted for each variant, we computed across-cohort heterogeneity chromosomes 1–22 and X data using the novel TOPMed P-values and I2 values using METAL. We considered a Freeze 5 on genome build GRCh38 and the Michigan P-value of 10−3 to indicate significant heterogeneity and Imputation Server (45). We provided the input data in used a significance threshold of P < 10−6 for joint P-values ‘vcf.gz’ format as GRCh38 build. We used the offered to determine suggestive statistical significance and the population panel ‘ALL’ and applied the server-side option common threshold of P < 5 × 10−8 to determine genome- to filter by an imputation R2 with threshold 0.1. The wide significance. final imputed results contained 80 794 511 variants in the Italian dataset, 75 346 562 variants in the Spanish Replication analysis of 13 release 5 COVID-19 dataset, 20 195 513 variants in the Norwegian dataset HGI associations and 53 164 135 variants in the German/Austrian dataset To replicate 13 genome-wide significant associations after TOPMed imputation. For the imputation of the reported by the COVID-19 HGI (7), we excluded 3837 X chromosome, we coded males as diploid in the individuals overlapping between our current study and non-pseudoautosomal (non-PAR) region. After QC and COVID-19 HGI release 5. We subsequently carried our imputation using TOPMed, in total, 8 910 172 variants logistic regression and meta-analysis as described above were included for the Italian panel, 9 089 877 variants for in 1579 cases and 10 372 population controls for the main the Spanish panel, 8 841 609 variants for the Norwegian analysis and 944 cases and 10 065 population controls panel and 9 019 898 variants for the German/Austrian for the subtype analysis (Supplementary Material, Table panel with post-imputation R2 ≥ 0.6 and MAF ≥ 1%. S1e). A threshold of 0.05 was used to define replicability of the results. Meta-analysis with the COVID-19 HGI Statistical analysis release 5 summary statistics included up to 12 888 Previous to the statistical analysis, we excluded qced hospitalized cases (including 5582 critically ill cases; individuals with a mild disease phenotype (did not COVID19_HGI_B2_ALL_leave_23andme_20210107.txt.gz; receive respiratory support) or missing age information. COVID19_HGI_A2_ALL_leave_23andme_20210107.txt.gz) This left 1536 Italian COVID-19 cases, 4734 Italian and 1 295 966 population controls of the COVID-19 HGI. controls, 1416 Spanish COVID-19 cases, 4382 Spanish We performed meta-analysis with METAL as described controls, 62 Norwegian cases, 262 Norwegian controls, above. To unify case definitions by our study and the 241 German COVID-19 cases and 3110 German controls COVID-19 HGI, we performed meta-analysis of summary for analysis (Supplementary Material, Table S1e). statistics from our main analysis with COVID-19 release 3960 | Human Molecular Genetics, 2022, Vol. 31, No. 23 5 A2 summary statistics and meta-analysis of our age∗ age, biological sex∗ age and the first 10 PCs from PCA subtype analysis and COVID-19 release 5 B2 summary as covariates dependent on the stratification level (i.e. for statistics. We used the common threshold of P < 5 × 10−8 sex-stratified analyses, only age and PCs were included; to determine genome-wide significance. for age-stratified analysis, only the biological sex and PCs were included; detailed formulas are shown in the Sup- Bayesian fine-mapping analysis plementary Methods). All analysis was done in R version Statistical fine-mapping analysis was conducted using 3.6.2. An inverse-variance weighted fixed-effects meta- FINEMAP (15) Version 1.4 for each of the loci of inter- analysis was conducted using the R-package metafor (47) est to calculate the posterior inclusion probability for to combine cohort-specific summary statistics, including each lead SNP and every other SNP within 250 kb flank- only statistics from cohorts with NCase and NControl > 50. ing regions. In the case where the association signal Age was analysed in groups ≤60 to >60, which were extended over a larger region, as in the case of 17q21.21, chosen in reference to the analyses performed by the the region was expanded to include this. FINEMAP deter- COVID-19 HGI. We additionally performed an interaction mined the 95% credible set of SNPs assuming a sin- analysis of SNP and sex as well as SNP and binarized age gle causal variant using shotgun stochastic search (—n- (≤60 versus >60), to determine statistical significance causal-snps 1—sss), i.e. the minimum set of variants con- of differences between the groups, respectively (Supple- taining the causal variant with ≥95% certainty. The union mentary Methods). For variants that showed a significant of the genotypes of the Italian, Spanish, Norwegian and age effect, we additionally analysed age groups 40–51, German/Austrian cohorts was used as the LD reference 51–60 and 61–70 for the Italian and Spanish cohorts and population. performed a meta-analysis on the summary statistics from these two cohorts only (Supplementary Material, Meta-analysis model-based assessment Fig. S10). We were interested in observing more detailed of replicability age effects in these age intervals with sufficient data for We tested the replicability of our candidate variants analysis. For the meta-analyses, P-values were adjusted from the meta-analysis using the Meta-Analysis Model- for multiple testing using the P-value correction of Ben- based Assessment of replicability (14) (MAMBA) method, jamini–Hochberg. as single outlier studies can drive a false positive meta- analysis association. MAMBA takes the genetic effects Imputation of the ∼0.9-Mb inversion and standard deviations from participating studies to polymorphism at 17q21.31 test each SNP for a true non-zero effect. This is achieved In 2005, Stefansson and colleagues discovered a 900 kb by fitting a two-level mixture model to genome-wide inversion polymorphism at 17q21.31, a region that LD-pruned SNPs that reduces to a fixed effect meta- contains several genes, including those encoding cor- analysis in absence of outliers, in which case it has ticotropin releasing hormone receptor 1 (CRHR1) and similar power. Since this is a likelihood model, the result microtubule-associated protein tau (MAPT) (17). Chro- is a posterior probability of replicability (PPR) that the mosomes with the inverted segment in different orien- SNP has a non-zero replicable effect. If the PPR is low, tations represent two distinct lineages, H1 and H2, that MAMBA tests for excess in variation of effects sizes and have diverged for up to 3 million years and show no outputs the probability that each study is an outlier. evidence of recombination (17). For the Italian, Spanish, For each of the first and second analyses, we used the Norwegian and German/Austrian GWAS discovery lead variants and a genome-wide set of LD-pruned SNPs cohorts, we inferred the 17q21.31 inversion status (H1 or (PLINK v1.90b6.16 64-bit, —indep-pairwise 500 kb 1 0.1) H2) with IMPUTE v2.3.2 (18) using genotype information (40,41) from each participating study for the algorithm and an imputation reference panel consisting of 109 to estimate the null distribution. Suggestive variants and individuals (from different continents [EUR, EAS, AFR]) LD-pruned background SNPs are entered as a single table from the 1000 Genomes Project Phase 3 (43), for which and the algorithm has no prior knowledge of which SNPs 17q21.31 inversion genotypes, obtained experimentally are considered significant. MAMBA was run with default by FISH (48,49) and droplet digital PCR (50), as well as settings and always terminated before 10 000 iterations. SNP genotype data are available for the region of the We considered a PPR value >0.8 to indicate a high prob- inversion. Imputed inversion genotypes were determined ability of replication. according to the highest posterior probability and were further confirmed by examining consensus genotypes of Association analysis of candidate SNPs known inversion tag SNPs in perfect LD (r2 = 1) with the We performed age- and sex-stratified on 7 candidate inversion in the imputation reference panel. H1 and H2 SNPs from our main and subtype analysis and 13 candi- alleles were coded according to an additive model as 0,1 date SNPs from the COVID-19 HGI analysis, as well as the and 2. novel variant identified in this study (7). We additionally We additionally imputed the 17q21.31 inversion for the analysed disease severity and comorbidities hyperten- COVID-19 HGI release 5 A2 and B2 analyses from the files sion, CAD and diabetes in cases only. We carried out a COVID19_HGI_A2_ALL_leave_23andme_20210107.txt.gz logistic regression analysis including age, biological sex, and COVID19_HGI_B2_ALL_leave_23andme_20210107.- Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3961 txt.gz. Here, the inversion P-value was imputed from genes that are located within locus boundaries or summary statistics using Fast and accurate P-value that are candidates based on GWAS colocalization Imputation for genome-wide association study (FAPI) with eQTL (eGenes) or sQTL (sGenes)-variants were (51). The odds ratio (OR) and 95% confidence intervals chosen. More precisely, the boundaries for the 19q13.33 (CIs) were estimated from the SNP rs62061809, which is locus were defined by Bayesian fine mapping (GRCh38: in perfect LD with the inversion (r2 = 1). The inversion was chr19:50344768–50379362), while extended boundaries coded as 0 for the major allele H1 and 1 for the minor (GRCh38: chr17:45495836–46707123) (54) were used for allele H2. All association analyses were carried out as the 17q21.21.31 locus, since it lies within a ∼0.9 Mb described above on the minor allele H2. inversion region of high LD that affects expression and splicing of numerous genes in GTEx (55). To retrieve candidate genes that overlap the boundaries, GENCODE Functional analysis v36 (56) annotations for genome build GRCh38 were used. Phenome-wide Association study (PheWAS) A complete list of candidate protein-coding genes from Phenotype associations with the different variants were both loci is provided in Supplementary Material, Table obtained from the NHGRI-EBI GWAS Catalog curated S21. collection of established genome-wide significant dis- ease/trait associations (http://www.ebi.ac.uk/gwas/; Gene expression analysis for candidate genes release 2020-12-02) (19). We queried GWAS hits in high from genome-wide significant susceptibility loci LD (r2 > 0.9) with the inversion as well as hits in high LD Publicly available bulk tissue and immune cell type RNA- (r2 > 0.9) with rs1405655 (19q13.33) and P-value < 10−7 . seq data for all available candidate genes were retrieved Since each GWAS study is focused on populations from from GTEx v8 (55) and from the Expression Atlas different origins, the LD patterns employed to evaluate (57) (BLUEPRINT consortium data [accession E-MTAB- the association between the inversion and GWAS signals 3827] (58)) portals, respectively. Gene-level expression were based on individuals with the corresponding values (transcripts per million, TPM) by tissue or by ancestry or the closest one available from our inversion cell type were obtained as median-summarized in the imputation panel (of 109 experimentally genotyped case of the GTEx data and as mean-summarized in individuals), whereas the global LD was selected if the case of the BLUEPRINT data. The summarized TPM populations from underrepresented ancestries were values were centered gene-wise and z-score scaled for studied .27 proxy variants were used for r1405655 and visualization using the ggplot2 R-package. The single-cell 2904 proxy variants were used for the inversion. RNA-seq (scRNA-seq) data in COVID-19 relevant tissues from non-diseased individuals, such as lung and upper Colocalization of GWAS and tissue-specific airways (59) or brain (60), were obtained from the COVID- expression and splicing quantitative trait loci 19 Cell Atlas (61). The pre-processed and cell type anno- (eQTL, sQTL) tated scRNA-seq datasets were retrieved as AnnData Colocalization of GWAS and QTL (eQTL and sQTL) data objects in.h5ad format files. Log-normalized average from 49 different human tissues of the GTEx Project expression values of available candidate genes by cell (GTEx Analysis Release V8) (19) was performed using the type were visualized using the scanpy v1.4.6 package HyPrColoc method (52) implemented in the ezQTL tool (62). For gene expression analysis of candidate genes (53). The analysis was conducted including all cis-QTL in SARS-CoV-2 infected brain organoids, pre-processed variants falling within ±75 kb from coordinates (GRCh38) and cell-annotated scRNA-seq data were obtained upon of the two variants of interest: chr17:46142465:T:A— request from Song et al. (63). Differential gene expression lead SNP for the 17q21.31 inversion—and chr19:50379362 analysis of scRNA-seq data was performed using the (rs1405655—lead SNP of the 19q13.33 locus). LD values R package MAST (64). More precisely, hurdle models used in the colocalization analysis were calculated based were used to evaluate differentially expressed genes on the 1000 Genomes population European (2) popula- in each brain organoid cell type (neural progenitors, tion only, to account for the ethnic bias of GTEx donors. interneurons, neurons and cortical neurons) comparing The loci of tested traits were considered to be colocalized SARS-CoV-2 infected and mock (non-infected) cells. when a regional probability of colocalization was greater The models were fitted using the condition, sample than 0.9. Additionally, to assess the actual effect (and its identification, number of detected genes (centered) and direction) of the variants on neighboring gene expression, total counts of SARS-CoV-2 transcripts (centered) as normalized effect score (NES) and P-values of the two covariates, thus adjusting for the cellular detection rate, variants (inversion tagged by rs62061809) were retrieved batch effects and viral load. Genes with PFDR < 0.01 and from the GTEx portal. absolute value of log2 fold change >0.1 were considered as significantly differentially expressed. Finally, the Selection and definition of candidate genes at status, log2 fold change and P-values of candidate gene 17q21.31 and 19q13.33 differential expression in COVID-19-infected lung cells To identify candidate genes most likely to play a causa- were obtained from pseudo-bulk differential expression tive role at 17q21.31 and 19q13.33, all protein-coding analysis performed by Delorey et al. (20) 3962 | Human Molecular Genetics, 2022, Vol. 31, No. 23 Analysis and fine mapping of the HLA as ‘no call’, individuals with dosages ≤1.3 were called Detailed descriptions of this analysis can be found in ‘secretors’ and individuals with dosages ≥1.7 were called the Supplementary Methods. In brief, quality-controlled ‘non-secretors’. Association analysis was carried out as genotypes at the HLA region (chr6:29-34 Mb) were described in the Section Statistical Analysis—Association extracted. HLA allele, amino acid and SNP imputation Analysis of Candidate SNPs on blood type or blood type were performed using the random-forest-based HLA secretor status coded as absent (0) or present (1). genotype imputation with attribute bagging (HIBAG) and applying specially tailored as well as publicly available Availability of summary statistics reference panels (5,65,66). The resulting data were used The GWAS summary statistics are available through as a basis for several subsequent analyses, including: (i) GWAS Catalog (https://www.ebi.ac.uk/gwas/), study basic association analysis (fine mapping) as described in accession number GCST90129579. the section Statistical Analysis and the Supplementary Methods, (ii) a peptidome-wide association study (11) (pepWAS), to screen for disease-relevant peptides from Supplementary Material SARS-CoV-2, that may present a possible functional Supplementary Material is available at HMG online. link between severe COVID-19 and variation at classical HLA loci, (iii) quantitative HLA analyses directed at the Acknowledgements number of peptides bound by an HLA allele as well as (iv) an analysis of HLA-presentation of shared peptides We would like to thank the COVID-19 Human Genetics (‘molecular mimicry’). Initiative (HGI) for collecting and openly sharing exten- sive summary statistics with the community. Analysis of the Y-chromosome haplotypes For the cohort from Milan Fondazione Ca’ Granda, we First, we produced high-quality Y-chromosome geno- would like to acknowledge specifically Rossana Carpani. types by manually calling and visually inspecting Y- A full list of the investigators who contributed to the chromosome SNPs in the male individuals from all generation of the GCAT data is available from www. cohorts only. Next, we used 22 Y-chromosome SNPs genomesforlife.com/. We thank Dr Luis Puig and Vanessa to distinguish known Y-chromosome haplogroups as Pleguezuelos on behalf of the Blood and Tissue Bank from described in the Supplementary Methods at different Catalonia (BST) who collaborated in the GCAT/COVICAT haplogroup resolutions. We here focused on haplogroup recruitment, and all the GCAT volunteers that partic- R, the most prevalent Y-chromosome haplogroup across ipated in the study. We would also like to acknowledge Europe. Association analysis was carried out as described the Task Force Humanitas and the Task Force Humanitas in the Section Statistical Analysis—Association Analysis Gavazzeni Castelli (Supplementary Note). We would like of Candidate SNPs on Y-chromosome haplogroups coded to acknowledge all the participants from the ‘Grupo de as absent (0) or present (1). We additionally analysed Trabajo en Medicina Personalizada contra el COVID-19 the end-point mortality (Supplementary Methods). The de Andalucia’. We also thank ‘Consejeria de Salud y coefficient of variation (CV) across frequencies was Familias’ and ‘Junta de Andalucía’ for their support and calculated for Y-chromosome haplogroups between funding that is currently supporting the COVID-19-GWAS individual batches (i.e. contributing center/hospital or and the COVID-PREMED initiatives. We are indebted different versions of the GSA). Here, only frequencies to the HCB-IDIBAPS Biobank for the biological human calculated from batches including more than 100 samples and data procurement and to the Fundació individuals were considered. Glòria Soler for its support to the COVIDBANK collection. We acknowledge the work of the Norwegian SARS-CoV-2 Analysis of the ABO secretor status Study group (Supplementary Note). We also acknowledge ABO blood group typing was performed as described Benedicte A. Lie at the Department of Immunology, by Ellinghaus et al. (5). Brief ly, genotypes of the SNPs Oslo University Hospital and the Norwegian Bone rs8176747 (c.803C > T, ABO∗ B), rs41302905 (c.802G > A, Marrow Donor Registry are acknowledged for providing ABO∗ O2) and rs8176719 (c.261delG, ABO∗ O1) were healthy controls. We thank the support of the German extracted from the imputed data (R2 = 1 for all SNPs COVID-19 multiomics initiative (decoi.eu). We thank the and cohorts) and used to infer the A, B and O blood members of the COVID-19 use and access committee types. The ABO-‘secretor’ status was inferred from the (Prof. Reinhold Förster, Prof. Christine Falk, Prof. Thomas genotypes of the rs601338 SNP (c.428G > A, FUT2∗ 01 N.02) Fühner, Prof. Christoph Höner zu Siederdissen, Prof. at the FUT2 gene, located at 19q13.33 and extracted Marius Höper, Prof. Thomas Friedrich Schulz, Prof. from the imputed data (R2 = 0.98–0.99 for all cohorts). Markus Cornberg, Prof. Thomas Illig and Prof. Michael Individuals carrying genotypes GA or GG were assigned Manns), as well as the study nurses, medical doctors secretor status and individuals carrying genotype AA and biobank staff members involved in the project. For were assigned non-secretor status based on the genotype funding, we thank the Ministry of Science and Culture dosages, ranging from 0 to 2, retrieved from the imputed (MWK) of Lower-Saxony. We thank everybody involved data. Individuals with allelic dosages 1.3–1.7 were called in the COVID-19 registry at the University Hospital Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3963 Regensburg (COVUR). Genotyping of the German control grants from Abbvie, Gilead Sciences and Intercept. dataset was performed at the Regeneron Genetics Center. J.C.H. received a philanthropic donation from Vivaldi Genotyping of the COMRI study was performed by Invest A/S owned by Jon Stephenson von Tetzchner the Genotyping laboratory of Institute for Molecular during the conduct of this study. C.D.S. reports grants, Medicine Finland FIMM Technology Centre, University of personal fees from AstraZeneca, personal fees and non- Helsinki. We are very grateful to Prof. Akiko Iwasaki and financial support from BBraun Melsungen, personal Eric Song for generously sharing pre-processed single- fees from BioNtech, grants, personal fees and non- cell RNA-seq data generated from SARS-CoV-2-infected financial support from Gilead Sciences, grants and brain organoids. personal fees from Janssen-Cilag, personal fees from Eli Lilly, personal fees from Formycon, personal fees Conf lict of Interest statement. P.K. has received lecture from Roche, other from Apeiron, grants and personal honoraria from or is advisor to Akademie für Infektion- fees from MSD, grants from Cepheid, personal fees from smedizin e.V., Astellas Pharma, European Confederation GSK, personal fees from Molecular partners, other from of Medical Mycology, Gilead Sciences, GPR Academy Eli Lilly, personal fees from SOBI during the conduct Ruesselsheim, MSD Sharp & Dohme GmbH, Noxxon of the study; personal fees from AbbVie, personal fees N.V. and University Hospital, LMU Munich outside the from Synairgen, grants and personal fees from ViiV submitted work. He also received nonfinancial scientific Healthcare, outside the submitted work. F.B. reports grants from Miltenyi Biotec GmbH. grants and personal fees from Astrazeneca, grants and O.A.C. reports grants or contracts from Amplyx, Basilea, personal fees from Chiesi, grants and personal fees BMBF, Cidara, DZIF, EU-DG RTD (101037867), F2G, Gilead, from GSK, personal fees from Grifols, personal fees Matinas, MedPace, MSD, Mundipharma, Octapharma, from Guidotti, personal fees from Insmed, grants and Pfizer, Scynexis; Consulting fees from Amplyx, Biocon, personal fees from Menarini, personal fees from Novartis, Biosys, Cidara, Da Volterra, Gilead, Matinas, Med- grants and personal fees from Pfizer, personal fees from Pace, Menarini, Molecular Partners, MSG-ERC, Noxxon, Zambon, personal fees from Vertex and personal fees Octapharma, PSI, Scynexis, Seres; Honoraria for lectures from Viatris outside the submitted work. C.L. reports from Abbott, Al-Jazeera Pharmaceuticals, Astellas, Grupo personal fees from Chiesi, Gilead, Janssen, Novartis, Biotoscana/United Medical/Knight, Hikma, MedScape, Oxfordimmunotec and Insmed outside the submitted MedUpdate, Merck/MSD, Mylan, Pfizer; Payment for work. J.H. reports personal fees from Chiesi, Gilead and expert testimony from Cidara; Participation on a Data Janssen outside the submitted work. T.H.K. has received Safety Monitoring Board or Advisory Board from Actelion, consulting fees from Engitix and Intercept and speaker Allecra, Cidara, Entasis, IQVIA, Jannsen, MedPace, fees from Gilead and Alfasigma. Paratek, PSI, Shionogi; a pending patent currently reviewed at the German Patent and Trade Mark Office; other interests from DGHO, DGI, ECMM, ISHAM, MSG- Funding ERC, Wiley. S.D. received funding from the Banca S.E.H. and C.A.S. partially supported genotyping through Intesa San Paolo. A.M. has received research funding a philanthropic donation. A.F. and D.E. were supported by from or reports personal fees from Dolce&Gabbana a grant from the German Federal Ministry of Education Fashion Firm, Ventana, Pierre Fabre, Verily, AbbVie, and COVID-19 grant Research (BMBF; ID:01KI20197); A.F., Astra Zeneca, Verseau Therapeutics, Compugen, Myeloid D.E. and F.D. were supported by the Deutsche Forschungs- Therapeutics, Third Rock Venture, Imcheck Therapeutics, gemeinschaft Cluster of Excellence ‘Precision Medicine in Ellipses, Novartis, Roche, Macrophage Pharma, Biove- Chronic Inflammation’ (EXC2167). D.E. was supported by locIta, Merck, Principia, Cedarlane Laboratories Ltd, the German Federal Ministry of Education and Research HyCult Biotechnology, eBioscience, Biolegend, ABCAM (BMBF) within the framework of the Computational Life Plc, Novus Biologicals, Enzo Life (ex Alexis Corp.), Sciences funding concept (CompLS grant 031L0165). D.E., Affymetrix, BMS, Johnson&Johnson; in addition, he has K.B. and S.B. acknowledge the Novo Nordisk Foundation a patent WO2019057780 ‘Anti-human migration stim- (NNF14CC0001 and NNF17OC0027594). T.L.L., A.T. and ulating factor (MSF) and uses thereof’ issued, a patent O.Ö. were funded by the Deutsche Forschungsgemein- WO2019081591 ‘NK or T cells and uses thereof’ issued, schaft (DFG, German Research Foundation), project a patent WO2020127471 ‘Use of SAP for the treatment numbers 279645989; 433116033; 437857095. M.W. and of Euromycetes fungi infections’ issued and a patent H.E. are supported by the German Research Foun- EP20182181.6 ‘PTX3 as prognostic marker in Covid-19’ dation (DFG) through the Research Training Group pending. A.L. has consulted by Intercept Pharma, Takeda, 1743, ‘Genes, Environment and Inflammation’. L.V. AlfaSigma, Abbvie, Gilead, and Merck Sharp, and Dohme. received funding from: Ricerca Finalizzata Ministero M.R.G. has served as a speaker for AbbVie, Bristol-Myers della Salute (RF-2016-02364358), Italian Ministry of Squibb, GENFIT, Gilead Sciences, Intercept, MSD and Health ‘CV PREVITAL’—strategie di prevenzione primaria Roche; an advisory board member for GENFIT, Gilead cardiovascolare primaria nella popolazione italiana; Sciences, Intercept, Janssen-Cilag, Kaleido, NovoNordisk, The European Union (EU) Programme Horizon 2020 Medimmune and Prosceinto; and has received research (under grant agreement No. 777377) for the project 3964 | Human Molecular Genetics, 2022, Vol. 31, No. 23 LITMUS- and for the project ‘REVEAL’; Fondazione IRCCS grants from Research Council of Norway grant no Ca’ Granda ‘Ricerca corrente’, Fondazione Sviluppo Ca’ 312780 during the conduct of the study. E.S. reports Granda ‘Liver-BIBLE’ (PR-0391), Fondazione IRCCS Ca’ grants from Research Council of Norway grant no. Granda ‘5permille’ ‘COVID-19 Biobank’ (RC100017A). A.B. 312769. The BioMaterialBank Nord is supported by was supported by a grant from Fondazione Cariplo to the German Center for Lung Research (DZL), Airway Fondazione Tettamanti: ‘Bio-banking of Covid-19 patient Research Center North (ARCN). The BioMaterialBank samples to support national and international research Nord is member of popgen 2.0 network (P2N). P.K. (Covid-Bank). This research was partly funded by an Bergisch Gladbach, Germany and the Cologne Excellence MIUR grant to the Department of Medical Sciences, under Cluster on Cellular Stress Responses in Aging-Associated the program ‘Dipartimenti di Eccellenza 2018–2022’. This Diseases, University of Cologne, Cologne, Germany. He is study makes use of data generated by the GCAT-Genomes supported by the German Federal Ministry of Education for Life. Cohort study of the Genomes of Catalonia, and Research (BMBF). O.A.C. is supported by the German Fundació IGTP (The Institute for Health Science Research Federal Ministry of Research and Education and is funded Germans Trias i Pujol) IGTP is part of the CERCA by the Deutsche Forschungsgemeinschaft (DFG, German Program/Generalitat de Catalunya. GCAT is supported Research Foundation) under Germany’s Excellence by Acción de Dinamización del ISCIII-MINECO and the Strategy—CECAD, EXC 2030–390661388. The COMRI Ministry of Health of the Generalitat of Catalunya (ADE cohort is funded by Technical University of Munich, 10/00026); the Agència de Gestió d’Ajuts Universitaris i de Munich, Germany. This work was supported by grants Recerca (AGAUR) (2017-SGR 529). M.M. received research of the Rolf M. Schwiete Stiftung, the Saarland University, funding from grant PI19/00335 Acción Estratégica en BMBF and The States of Saarland and Lower Saxony. Salud, integrated in the Spanish National RDI Plan and K.U.L. is supported by the German Research Foundation financed by ISCIII-Subdirección General de Evaluación (DFG, LU-1944/3-1). Genotyping for the BoSCO study is and the Fondo Europeo de Desarrollo Regional (European funded by the Institute of Human Genetics, University Regional Development Fund (FEDER)-Una manera de Hospital Bonn. F.H. was supported by the Bavarian State hacer Europa’). B.C. is supported by national grants Ministry for Science and Arts. Part of the genotyping PI18/01512. X.F. is supported by the VEIS project (001- was supported by a grant to A.R. from the German P-001647) (co-funded by the European Regional Develop- Federal Ministry of Education and Research (BMBF, ment Fund (ERDF), ‘A way to build Europe’). Additional grant: 01ED1619A, European Alzheimer DNA BioBank, data included in this study were obtained in part by EADB) within the context of the EU Joint Programme— the COVICAT Study Group (Cohort Covid de Catalunya) Neurodegenerative Disease Research (JPND). Additional supported by IsGlobal and IGTP, European Institute of funding was derived from the German Research Foun- Innovation & Technology (EIT), a body of the European dation (DFG) grant: RA 1971/6-1 to A.R. P.R. is supported Union, COVID-19 Rapid Response activity 73A and SR20- by the DFG (CCGA Sequencing Centre and DFG ExC2167 01024 La Caixa Foundation. A.J. and S.M. were supported PMI and by SH state funds for COVID19 research). F.T. by the Spanish Ministry of Economy and Competitiveness is supported by the Clinician Scientist Program of the (grant numbers: PSE-010000-2006-6 and IPT-010000- Deutsche Forschungsgemeinschaft Cluster of Excellence 2010-36). A.J. was also supported by national grant ‘Precision Medicine in Chronic Inflammation’ (EXC2167). PI17/00019 from the Acción Estratégica en Salud (ISCIII) C.L. and J.H. are supported by the German Center and the European Regional Development Fund (FEDER). for Infection Research (DZIF). T.B., M.M.B., O.W. und The Basque Biobank, a hospital-related platform that A.H. are supported by the Stiftung Universitätsmedizin also involves all Osakidetza health centres, the Basque Essen. M.A.-H. was supported by Juan de la Cierva government’s Department of Health and Onkologikoa, is Incorporacion program, grant IJC2018-035131-I funded operated by the Basque Foundation for Health Innovation by MCIN/AEI/10.13039/501100011033. E.C.S. is supported and Research-BIOEF. M.C. received Grants BFU2016- by the Deutsche Forschungsgemeinschaft (DFG; SCHU 77244-R and PID2019-107836RB-I00 funded by the Agen- 2419/2-1). cia Estatal de Investigación (AEI, Spain) and the European Regional Development Fund (FEDER, EU). M.R.G., J.A.H., Authors’ contributions R.G.D. and D.M.M. are supported by the ‘Spanish Ministry Contributions are shown in the Supplementary Note. of Economy, Innovation and Competition, the Instituto de Salud Carlos III’ (PI19/01404, PI16/01842, PI19/00589, References PI17/00535 and GLD19/00100) and by the Andalussian 1. Booth, A., Reed, A.B., Ponzo, S., Yassaee, A., Aral, M., Plans, D., government (Proyectos Estratégicos-Fondos Feder PE- Labrique, A. and Mohan, D. (2021) Population risk factors for 0451-2018, COVID-Premed, COVID GWAs). The position severe disease and mortality in COVID-19: a global systematic held by Itziar de Rojas Salarich is funded by grant review and meta-analysis. PLoS One, 16, e0247461. FI20/00215, PFIS Contratos Predoctorales de Formación 2. Wong, C.K.H., Wong, J.Y.H., Tang, E.H.M., Au, C.H. and Wai, en Investigación en Salud. Enrique Calderón’s team is A.K.C. (2020) Clinical presentations, laboratory and radiological supported by CIBER of Epidemiology and Public Health findings, and treatments for 11,028 COVID-19 patients: a sys- (CIBERESP), ‘Instituto de Salud Carlos III’. J.C.H. reports tematic review and meta-analysis. Sci. Rep., 10, 19765. Human Molecular Genetics, 2022, Vol. 31, No. 23 | 3965 3. Peckham, H., de Gruijter, N.M., Raine, C., Radziszewska, A., new NHGRI-EBI catalog of published genome-wide association Ciurtin, C., Wedderburn, L.R., Rosser, E.C., Webb, K. and Deakin, studies (GWAS catalog). Nucleic Acids Res., 45, D896–D901. C.T. (2020) Male sex identified by global COVID-19 meta-analysis 20. Delorey, T.M., Ziegler, C.G.K., Heimberg, G., Normand, R., Yang, Y., as a risk factor for death and ITU admission. Nat. Commun., 11, Segerstolpe, Å., Abbondanza, D., Fleming, S.J., Subramanian, A., 6317. Montoro, D.T. et al. (2021) COVID-19 tissue atlases reveal SARS- 4. Hu, Y., Sun, J., Dai, Z., Deng, H., Li, X., Huang, Q., Wu, Y., Sun, L. CoV-2 pathology and cellular targets. Nature, 595, 107–113. and Xu, Y. (2020) Prevalence and severity of corona virus disease 21. Quach, H., Rotival, M., Pothlichet, J., Loh, Y.H.E., Dannemann, 2019 (COVID-19): a systematic review and meta-analysis. J. Clin. M., Zidane, N., Laval, G., Patin, E., Harmant, C., Lopez, M. Virol., 127, 104371. et al. (2016) Genetic adaptation and Neandertal admixture 5. Severe Covid-19 GWAS Group (2020) Genomewide association shaped the immune system of human populations. Cell, 167, study of severe Covid-19 with respiratory failure. N. Engl. J. Med., 643–656.e17. 383, 1522–1534. 22. Mankelow, T.J., Singleton, B.K., Moura, P.L., Stevens-Hernandez, 6. Pairo-Castineira, E., Clohisey, S., Klaric, L., Bretherick, A.D., Raw- C.J., Cogan, N.M., Gyorffy, G., Kupzig, S., Nichols, L., Asby, C., Poo- lik, K., Pasko, D., Walker, S., Parkinson, N., Fourman, M.H., Russell, ley, J. et al. (2021) Blood group type a secretors are associated with C.D. et al. (2021) Genetic mechanisms of critical illness in COVID- a higher risk of COVID-19 cardiovascular disease complications. 19. Nature, 591, 92–98. eJHaem, 2, 175–187. 7. COVID-19 Host Genetics Initiative, C (2021) Mapping the human 23. Hammer, M. (2002) A nomenclature system for the tree of genetic architecture of COVID-19. Nature, 600, 472–477. human Y-chromosomal binary haplogroups. Genome Res., 12, 8. Velavan, T.P., Pallerla, S.R., Rüter, J., Augustin, Y., Kremsner, P.G., 339–348. Krishna, S. and Meyer, C.G. (2021) Host genetic factors deter- 24. Jiang, X., Holmes, C. and McVean, G. (2021) The impact of age on mining COVID-19 susceptibility and severity. EBioMedicine, 72, genetic risk for common diseases. PLoS Genet., 17, e1009723. 103629. 25. Ward, H.E. and Nicholas, T.E. (1984) Alveolar type I and type II 9. Namkoong, H., Edahiro, R., Fukunaga, K., Shirai, Y., Sonehara, cells. Aust. NZ J. Med., 14, 731–734. K., Tanaka, H., Lee, H., Hasegawa, T., Kanai, M., Naito, 26. Melms, J.C., Biermann, J., Huang, H., Wang, Y., Nair, A., Tagore, S., T. et al. (2021) Japan COVID-19 task force: a nation- Katsyv, I., Rendeiro, A.F., Amin, A.D., Schapiro, D. et al. (2021) A wide consortium to elucidate host genetics of COVID- molecular single-cell lung atlas of lethal COVID-19. Nature, 595, 19 pandemic in Japan. medRxiv. 2021.05.17.21256513; 114–119. https://doi.org/10.1101/2021.05.17.2125651. 27. Islam, A.B.M.M.K. and Khan, M.A.A.K. (2020) Lung transcriptome 10. Li, Y., Ke, Y., Xia, X., Wang, Y., Cheng, F., Liu, X., Jin, X., Li, B., Xie, C., of a COVID-19 patient and systems biology predictions suggest Liu, S. et al. (2021) Genome-wide association study of COVID-19 impaired surfactant production which may be druggable by severity among the Chinese population. Cell Discov., 7, 76. surfactant therapy. Sci. Rep., 10, 19395. 11. Arora, J., McLaren, P.J., Chaturvedi, N., Carrington, M., Fellay, J. and 28. Wu, L., Zhu, J., Liu, D., Sun, Y. and Wu, C. (2021) An integrative Lenz, T.L. (2019) HIV peptidome-wide association study reveals multiomics analysis identifies putative causal genes for COVID- patient-specific epitope repertoires associated with HIV control. 19 severity. Genet. Med., 23, 2076–2086. Proc. Natl. Acad. Sci. U. S. A., 116, 944–949. 29. American Journal of Respiratory and Critical Care Medicine; 203, 12. Zhou, W., Nielsen, J.B., Fritsche, L.G., Dey, R., Gabrielsen, M.E., 2021. This seems to be from an online abstract submission. Wolford, B.N., LeFaive, J., VandeHaar, P., Gagliano, S.A., Gifford, 203.1_MeetingAbstracts.A3765. A. et al. (2018) Efficiently controlling for case-control imbalance 30. Astle, W.J., Elding, H., Jiang, T., Allen, D., Ruklisa, D., Mann, A.L., and sample relatedness in large-scale genetic association stud- Mead, D., Bouman, H., Riveros-Mckay, F., Kostadima, M.A. et al. ies. Nat. Genet., 50, 1335–1341. (2016) The allelic landscape of human blood cell trait variation 13. Willer, C.J., Li, Y. and Abecasis, G.R. (2010) METAL: fast and and links to common complex disease. Cell, 167, 1415–1429.e19. efficient meta-analysis of genomewide association scans. Bioin- 31. Sakornsakolpat, P., Prokopenko, D., Lamontagne, M., Reeve, N.F., formatics, 26, 2190–2191. Guyatt, A.L., Jackson, V.E., Shrine, N., Qiao, D., Bartz, T.M., Kim, 14. McGuire, D., Jiang, Y., Liu, M., Weissenkampen, J.D., Eckert, S., D.K. et al. (2019) Genetic landscape of chronic obstructive pul- Yang, L., Chen, F., Liu, M., Jiang, Y., Wedow, R. et al. (2021) Model- monary disease identifies heterogeneous cell-type and pheno- based assessment of replicability for genome-wide association type associations. Nat. Genet., 51, 494–505. meta-analysis. Nat. Commun., 12, 1964. 32. Tantisira, K.G., Lazarus, R., Litonjua, A.A., Klanderman, B. and 15. Benner, C., Spencer, C.C.A., Havulinna, A.S., Salomaa, V., Ripatti, Weiss, S.T. (2008) Chromosome 17: association of a large inver- S. and Pirinen, M. (2016) FINEMAP: efficient variable selection sion polymorphism with corticosteroid response in asthma. using summary data from genome-wide association studies. Pharmacogenet. Genomics, 18, 733–737. Bioinformatics, 32, 1493–1501. 33. den Dekker, A.D., Davis, F.M., Joshi, A.D., Wolf, S.J., Allen, R., Lipin- 16. Nakanishi, T., Pigazzini, S., Degenhardt, F., Cordioli, M., Butler- ski, J., Nguyen, B., Kirma, J., Nycz, D., Bermick, J. et al. (2020) TNF- Laporte, G., Maya-Miles, D., Nafría-Jiménez, B., Bouysran, Y., α regulates diabetic macrophage function through the histone Niemi, M., Palom, A. et al. (2021) Age-dependent impact of the acetyltransferase MOF. JCI Insight, 5, e132306. major common genetic risk factor for COVID-19 on severity and 34. Cursano, S., Battaglia, C.R., Urrutia-Ruiz, C., Grabrucker, S., mortality. J. Clin. Invest., 131, e152386. Schön, M., Bockmann, J., Braumüller, S., Radermacher, P., Roselli, 17. Stefansson, H., Helgason, A., Thorleifsson, G., Steinthorsdottir, F., Huber-Lang, M. et al. (2020) A CRHR1 antagonist pre- V., Masson, G., Barnard, J., Baker, A., Jonasdottir, A., Ingason, vents synaptic loss and memory deficits in a trauma-induced A., Gudnadottir, V.G. et al. (2005) A common inversion under delirium-like syndrome. Mol. Psychiatry., 26, 3778–3794. selection in Europeans. Nat. Genet., 37, 129–137. 35. Thompson, S.B., Sandor, A.M., Lui, V., Chung, J.W., Waldman, 18. Howie, B., Marchini, J. and Stephens, M. (2011) Genotype imputa- M.M., Long, R.A., Estin, M.L., Matsuda, J.L., Friedman, R.S. and tion with thousands of genomes. G3, 1, 457–470. Jacobelli, J. (2020) Formin-like 1 mediates effector t cell traffick- 19. MacArthur, J., Bowler, E., Cerezo, M., Gil, L., Hall, P., Hastings, E., ing to inflammatory sites to enable t cell-mediated autoimmu- Junkins, H., McMahon, A., Milano, A., Morales, J. et al. (2017) The nity. elife, 9. 10.7554/eLife.58046. 3966 | Human Molecular Genetics, 2022, Vol. 31, No. 23 36. Pickett, B.E., Greer, D.S., Zhang, Y., Stewart, L., Zhou, L., Sun, G., 53. Zhang, T., Klein, A., Sang, J., Choi, J. and Brown, K.M. (2022) Gu, Z., Kumar, S., Zaremba, S., Larsen, C.N. et al. (2012) Virus ezQTL: a web platform for interactive visualization and pathogen database and analysis resource (ViPR): a compre- colocalization of quantitative trait loci and GWAS. Genomics hensive bioinformatics database and analysis resource for the Proteomics Bioinformatics. 25:S1672-0229(22)00069-9. doi: coronavirus research community. Viruses, 4, 3209–3226. 10.1016/j.gpb.2022.05.004. Epub ahead of print. PMID: 35643189. 37. Kirkpatrick, M. (2010) How and why chromosome inversions 54. Martínez-Fundichely, A., Casillas, S., Egea, R., Ràmia, M., Bar- evolve. PLoS Biol., 8, e1000501. badilla, A., Pantano, L., Puig, M. and Cáceres, M. (2014) 38. Wellenreuther, M. and Bernatchez, L. (2018) Eco-evolutionary InvFEST, a database integrating information of polymorphic genomics of chromosomal inversions. Trends Ecol. Evol., 33, inversions in the human genome. Nucleic Acids Res., 42, 427–440. D1027–D1032. 39. Ben Shachar, S., Barda, N., Manor, S., Israeli, S., Dagan, N., Carmi, 55. Aguet, F., Barbeira, A.N., Bonazzola, R., Brown, A., Castel, S.E., Jo, S., Balicer, R., Zisser, B. and Louzoun, Y. (2021) MHC Haplotyping B., Kasela, S., Kim-Hellmuth, S., Liang, Y., Oliva, M. et al. (2020) of SARS-CoV-2 patients: HLA subtypes are not associated with The GTEx consortium atlas of genetic regulatory effects across the presence and severity of COVID-19 in the Israeli population. human tissues. Science (80-.), 369, 1318–1330. J. Clin. Immunol., 41, 1154–1161. 56. Frankish, A., Diekhans, M., Ferreira, A.M., Johnson, R., Jungreis, 40. Purcell, S. PLINK 1.9. PLINK 1.9. http://pngu.mgh.harvard.edu/ I., Loveland, J., Mudge, J.M., Sisu, C., Wright, J., Armstrong, J. purcell/plink/. et al. (2019) GENCODE reference annotation for the human and 41. Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M.A., mouse genomes. Nucleic Acids Res., 47, D766–D773. Bender, D., Maller, J., Sklar, P., de Bakker, P.I., Daly, M.J. and 57. Papatheodorou, I., Moreno, P., Manning, J., Fuentes, A.M.P., Sham, P.C. (2007) PLINK: a tool set for whole-genome association George, N., Fexova, S., Fonseca, N.A., Füllgrabe, A., Green, M., and population-based linkage analyses. Am. J. Hum. Genet., 81, Huang, N. et al. (2020) Expression atlas update: from tissues to 559–575. single cells. Nucleic Acids Res., 48, D77–D83. 42. Price, A.L., Weale, M.E., Patterson, N., Myers, S.R., Need, A.C., 58. Stunnenberg, H.G., Abrignani, S., Adams, D., de Almeida, M., Shianna, K.V., Ge, D., Rotter, J.I., Torres, E., Taylor, K.D.D. et al. Altucci, L., Amin, V., Amit, I., Antonarakis, S.E., Aparicio, S., Arima, (2008) Long-range LD can confound genome scans in admixed T. et al. (2016) The international human epigenome consortium: populations. Am. J. Hum. Genet., 83, 132–135. a Blueprint for scientific collaboration and discovery. Cell, 167, 43. Auton, A., Brooks, L.D., Durbin, R.M., Garrison, E.P., Kang, H.M., 1145–1149. Korbel, J.O., Marchini, J.L., McCarthy, S., McVean, G.A. and Abeca- 59. Vieira Braga, F.A., Kar, G., Berg, M., Carpaij, O.A., Polanski, K., sis, G.R. (2015) A global reference for human genetic variation. Simon, L.M., Brouwer, S., Gomes, T., Hesse, L., Jiang, J. et al. (2019) Nature, 526, 68–74. A cellular census of human lungs identifies novel cell states in 44. Abraham, G., Qiu, Y. and Inouye, M. (2017) FlashPCA2: principal health and in asthma. Nat. Med., 25, 1153–1163. component analysis of biobank-scale genotype datasets. Bioin- 60. Habib, N., Avraham-Davidi, I., Basu, A., Burks, T., Shekhar, K., formatics, 33, 2776–2778. Hofree, M., Choudhury, S.R., Aguet, F., Gelfand, E., Ardlie, K. et al. 45. Taliun, D., Harris, D.N., Kessler, M.D., Carlson, J., Szpiech, Z.A., (2017) Massively parallel single-nucleus RNA-seq with DroNc- Torres, R., Taliun, S.A.G., Corvelo, A., Gogarten, S.M., Kang, H.M. seq. Nat. Methods, 14, 955–958. et al. (2021) Sequencing of 53,831 diverse genomes from the 61. Sungnak, W., Huang, N., Bécavin, C., Berg, M., Queen, R., Litv- NHLBI TOPMed program. Nature, 590, 290–299. inukova, M., Talavera-López, C., Maatz, H., Reichart, D., Sampazi- 46. König, I.R., Loley, C., Erdmann, J. and Ziegler, A. (2014) How to otis, F. et al. (2020) SARS-CoV-2 entry factors are highly expressed include chromosome X in your genome-wide association study. in nasal epithelial cells together with innate immune genes. Nat. Genet. Epidemiol., 38, 97–103. Med., 26, 681–687. 47. Viechtbauer, W. (2010) Conducting meta-analyses in R with the 62. Wolf, F.A., Angerer, P. and Theis, F.J. (2018) SCANPY: large-scale metafor. J. Stat. Softw., 36, 1–48. single-cell gene expression data analysis. Genome Biol., 19, 15. 48. Antonacci, F., Kidd, J.M., Marques-Bonet, T., Ventura, M., Siswara, 63. Song, E., Zhang, C., Israelow, B., Lu-Culligan, A., Prado, A.V., P., Jiang, Z. and Eichler, E.E. (2009) Characterization of six human Skriabine, S., Lu, P., El Weizman, O., Liu, F., Dai, Y. et al. (2021) disease-associated inversion polymorphisms. Hum. Mol. Genet., Neuroinvasion of SARS-CoV-2 in human and mouse brain. J. Exp. 18, 2555–2566. Med., 218, e20202135. doi: 10.1084/jem.20202135. 49. Steinberg, K.M., Antonacci, F., Sudmant, P.H., Kidd, J.M., Camp- 64. Finak, G., McDavid, A., Yajima, M., Deng, J., Gersuk, V., Shalek, bell, C.D., Vives, L., Malig, M., Scheinfeldt, L., Beggs, W., Ibrahim, A.K., Slichter, C.K., Miller, H.W., McElrath, M.J., Prlic, M., Linsley, M. et al. (2012) Structural diversity and African origin of the P.S. and Gottardo, R. (2015) MAST: a flexible statistical frame- 17q21.31 inversion polymorphism. Nat. Genet., 44, 872–880. work for assessing transcriptional changes and characterizing 50. Puig, M., Lerga-Jaso, J., Giner-Delgado, C., Pacheco, S., Izquierdo, heterogeneity in single-cell RNA sequencing data. Genome Biol., D., Delprat, A., Gayà-Vidal, M., Regan, J.F., Karlin-Neumann, G. 16, 278. and Cáceres, M. (2020) Determining the impact of uncharacter- 65. Degenhardt, F., Wendorff, M., Wittig, M., Ellinghaus, E., Datta, ized inversions in the human genome by droplet digital PCR. L.W., Schembri, J., Ng, S.C., Rosati, E., Hübenthal, M., Ellinghaus, Genome Res., 30, 724–735. D. et al. (2019) Construction and benchmarking of a multi-ethnic 51. Kwan, J.S.H., Li, M.X., Deng, J.E. and Sham, P.C. (2016) FAPI: fast reference panel for the imputation of HLA class I and II alleles. and accurate P-value imputation for genome-wide association Hum. Mol. Genet., 28, 2078–2092. study. Eur. J. Hum. Genet., 24, 761–766. 66. Degenhardt, F., Mayr, G., Wendorff, M., Boucher, G., Ellinghaus, E., 52. Foley, C.N., Staley, J.R., Breen, P.G., Sun, B.B., Kirk, P.D.W., Burgess, Ellinghaus, D., ElAbd, H., Rosati, E., Hübenthal, M., Juzenas, S. et al. S. and Howson, J.M.M. (2021) A fast and efficient colocalization (2021) Trans-ethnic analysis of the human leukocyte antigen algorithm for identifying shared genetic risk factors across mul- region for ulcerative colitis reveals common disease signatures. tiple traits. Nat. Commun., 12, 764. Hum. Mol. Genet., 30, 356–369.