This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Hauberg ME, Roussos P, Grove J, Børglum AD, Mattheisen M; Schizophrenia Working Group of the Psychiatric Genomics Consortium. Analyzing the role of microRNAs in schizophrenia in the context of common genetic risk variants. JAMA Psychiatry. Published online March 9, 2016. doi:10.1001/jamapsychiatry.2015.3018.
eAppendix 1. Members of the Schizophrenia Working Group of the Psychiatric Genomics Consortium eAppendix 2. Supplemental Text eReferences eFigure 1. Circos plot for the MHC-region. eFigure 2. Clustering of the top-10 scoring miRNA families. eFigure 3. Region plots of GWAS associations for the miR-9-5p host genes. eFigure 4. Spatiotemporal brain expression of MIR137, MIR2682, and MIR9-2. eFigure 5. PPI-network of the miR-9-5p targetome. eFigure 6. Venn diagrams of the miR-9-5p/miR-137 and miR-137/-miR-2682-5p targetomes. eTable 1. Full results of the TargetScan conserved miRNA gene set analysis. eTable 2. Alternative TargetScan gene set analysis: longer clumping range. eTable 3. Alternative TargetScan gene set analysis: lower LD-threshold. eTable 4. Alternative TargetScan gene set analysis: lower LD-threshold and longer clumping range. eTable 5. Validation of top-10 gene sets using MiRanda target predictions. eTable 6. Validation of top-10 gene sets using TargetMiner target predictions. eTable 7. Results for the top-10 gene sets filtered with CLIP-data from 58 experiments. eTable 8. Signal for the top-10 gene sets in three unrelated traits. eTable 9. Functional annotation of genes regulated by the top-10 miRNA – biological processes. eTable 10. Functional annotation of genes regulated by the top-10 miRNA – annotation clusters. eTable 11. The top 50 most schizophrenia-associated genes targeted by miR-9-5p. eTable 12. Functional annotation of genes regulated by miR-9-5p – biological processes. eTable 13. Functional annotation of genes regulated by miR-9-5p – annotation clusters. eTable 14. Functional annotation of genes regulated by miR-9-5p – transcription factors. eTable 15. Genes in the schizophrenia enriched miR-9-5p subset. eTable 16. Gene set results for miRNA in schizophrenia GWAS loci. eTable 17. Gene set results for miRNA in schizophrenia CNVs.
This supplementary material has been provided by the authors to give readers additional information about their work.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
eAppendix 1. Members of the Schizophrenia Working Group of the Psychiatric Genomics Consortium Stephan Ripke1,2, Benjamin M. Neale1,2,3,4, Aiden Corvin5, James T. R. Walters6, Kai-How Farh1, Peter A. Holmans6,7, Phil Lee1,2,4, Brendan Bulik-Sullivan1,2, David A. Collier8,9, Hailiang Huang1,3, Tune H. Pers3,10,11, Ingrid Agartz12,13,14, Esben Agerbo15,16,17, Margot Albus18, Madeline Alexander19, Farooq Amin20,21, Silviu A. Bacanu22, Martin Begemann23, Richard A Belliveau Jr2, Judit Bene24,25, Sarah E. Bergen 2,26, Elizabeth Bevilacqua2, Tim B Bigdeli 22, Donald W. Black27, Richard Bruggeman28, Nancy G. Buccola29, Randy L. Buckner30,31,32, William Byerley33, Wiepke Cahn34, Guiqing Cai35,36, Murray J. Cairns39,120,170, Dominique Campion37, Rita M. Cantor38, Vaughan J. Carr39,40, Noa Carrera6, Stanley V. Catts39,41, Kimberly D. Chambert2, Raymond C. K. Chan42, Ronald Y. L. Chen43, Eric Y. H. Chen43,44, Wei Cheng45, Eric F. C. Cheung46, Siow Ann Chong47, C. Robert Cloninger48, David Cohen49, Nadine Cohen50, Paul Cormican5, Nick Craddock6,7, James J. Crowley51, David Curtis52,53, Michael Davidson54, Kenneth L. Davis36, Franziska Degenhardt55,56, Jurgen Del Favero57, Lynn E. DeLisi128,129 , Ditte Demontis17,58,59, Dimitris Dikeos60, Timothy Dinan61, Srdjan Djurovic14,62, Gary Donohoe5,63, Elodie Drapeau36, Jubao Duan64,65, Frank Dudbridge66, Naser Durmishi67, Peter Eichhammer68, Johan Eriksson69,70,71, Valentina Escott-Price6, Laurent Essioux72, Ayman H. Fanous73,74,75,76, Martilias S. Farrell51, Josef Frank77, Lude Franke78, Robert Freedman79, Nelson B. Freimer80, Marion Friedl81, Joseph I. Friedman36, Menachem Fromer1,2,4,82, Giulio Genovese2, Lyudmila Georgieva6, Elliot S. Gershon209, Ina Giegling81,83, Paola Giusti-Rodríguez51, Stephanie Godard84, Jacqueline I. Goldstein1,3, Vera Golimbet85, Srihari Gopal86, Jacob Gratten87, Lieuwe de Haan88, Christian Hammer23, Marian L. Hamshere6, Mark Hansen89, Thomas Hansen17,90, Vahram Haroutunian36,91,92, Annette M. Hartmann81, Frans A. Henskens39,93,94, Stefan Herms55,56,95, Joel N. Hirschhorn3,11,96, Per Hoffmann55,56,95, Andrea Hofman55,56, Mads V. Hollegaard97, David M. Hougaard97, Masashi Ikeda98, Inge Joa99, Antonio Julià100, René S. Kahn34, Luba Kalaydjieva101,102, Sena Karachanak-Yankova103, Juha Karjalainen78, David Kavanagh6, Matthew C. Keller104, Brian J. Kelly120, James L. Kennedy105,106,107, Andrey Khrunin108, Yunjung Kim51, Janis Klovins109, James A. Knowles110, Bettina Konte81, Vaidutis Kucinskas111, Zita Ausrele Kucinskiene111, Hana Kuzelova-Ptackova112, Anna K. Kähler26, Claudine Laurent19,113, Jimmy Lee Chee Keong47,114, S. Hong Lee87, Sophie E. Legge6, Bernard Lerer115, Miaoxin Li43,44,116 Tao Li117, Kung-Yee Liang118, Jeffrey Lieberman119, Svetlana Limborska108, Carmel M. Loughland39,120, Jan Lubinski121, Jouko Lönnqvist122, Milan Macek Jr112, Patrik K. E. Magnusson26, Brion S. Maher123, Wolfgang Maier124, Jacques Mallet125, Sara Marsal100, Manuel Mattheisen17,58,59,126, Morten Mattingsdal14,127, Robert W. McCarley128,129, Colm McDonald130, Andrew M. McIntosh131,132, Sandra Meier77, Carin J. Meijer88, Bela Melegh24,25, Ingrid Melle14,133, Raquelle I. Mesholam-Gately128,134, Andres Metspalu135, Patricia T. Michie39,136, Lili Milani135, Vihra Milanova137, Younes Mokrab8, Derek W. Morris5,63, Ole Mors17,58,138, Kieran C. Murphy139, Robin M. Murray140, Inez Myin-Germeys141, Bertram Müller-Myhsok142,143,144, Mari Nelis135, Igor Nenadic145, Deborah A. Nertney146, Gerald Nestadt147, Kristin K. Nicodemus148, Liene Nikitina-Zake109, Laura Nisenbaum149, Annelie Nordin150, Eadbhard O’Callaghan151, Colm O’Dushlaine2, F. Anthony O’Neill152, Sang-Yun Oh153, Ann Olincy79, Line Olsen17,90, Jim Van Os141,154, Psychosis Endophenotypes International Consortium155, Christos Pantelis39,156, George N. Papadimitriou60, Sergi Papiol23, Elena Parkhomenko36, Michele T. Pato110, Tiina Paunio157,158, Milica Pejovic-Milovancevic159, Diana O. Perkins160, Olli Pietiläinen158,161, Jonathan Pimm53, Andrew J. Pocklington6, John Powell140, Alkes Price3,162, Ann E. Pulver147, Shaun M. Purcell82, Digby Quested163, Henrik B. Rasmussen17,90, Abraham Reichenberg36, Mark A. Reimers164, Alexander L. Richards6, Joshua L. Roffman30,32, Panos Roussos82,165, Douglas M. Ruderfer6,82, Veikko Salomaa71, Alan R. Sanders64,65, Ulrich Schall39,120, Christian R. Schubert166, Thomas G. Schulze77,167, Sibylle G. Schwab168, Edward M. Scolnick2, Rodney J. Scott39,169,170, Larry J. Seidman128,134, Jianxin Shi171, Engilbert Sigurdsson172, Teimuraz Silagadze173, Jeremy M. Silverman36,174, Kang Sim47, Petr Slominsky108, Jordan W. Smoller2,4, Hon-Cheong So43, Chris C. A. Spencer175, Eli A. Stahl3,82, Hreinn Stefansson176, Stacy Steinberg176, Elisabeth Stogmann177, Richard E. Straub178, Eric Strengman179,34, Jana Strohmaier77, T. Scott Stroup119, Mythily Subramaniam47, Jaana Suvisaari122, Dragan M. Svrakic48, Jin P. Szatkiewicz51, Erik Söderman12, Srinivas Thirumalai180, Draga Toncheva103, Paul A. Tooney39,120,170 , Sarah Tosato181, Juha Veijola182,183, John Waddington184, Dermot Walsh185, Dai Wang86, Qiang Wang117, Bradley T. Webb22, Mark Weiser54, Dieter B. Wildenauer186, Nigel M. Williams6, Stephanie Williams51, Stephanie H. Witt77, Aaron R. Wolen164, Emily H. M. Wong43, Brandon K. Wormley22, Jing Qin Wu39,170, Hualin Simon Xi187, Clement C. Zai105,106, Xuebin Zheng188, Fritz Zimprich177, Naomi R. Wray87, Kari Stefansson176, Peter M. Visscher87, Wellcome Trust Case-Control Consortium 2189, Rolf Adolfsson150, Ole A. Andreassen14,133, Douglas H. R. Blackwood132, Elvira Bramon190, Joseph D. Buxbaum35,36,91,191, Anders D. Børglum17,58,59,138, Sven Cichon55,56,95,192, Ariel Darvasi193, Enrico Domenici194, Hannelore Ehrenreich23, Tõnu Esko3,11,96,135, Pablo V. Gejman64,65, Michael Gill5, Hugh Gurling53, Christina M. Hultman26, Nakao Iwata98, Assen V. Jablensky39,102,186,195, Erik G. Jönsson12,14, Kenneth S. Kendler196, George Kirov6, Jo Knight105,106,107, Todd Lencz197,198,199, Douglas F. Levinson19, Qingqin S. Li86, Jianjun Liu188,200, Anil K. Malhotra197,198,199, Steven A. McCarroll2,96, Andrew McQuillin53, Jennifer L. Moran2, Preben B. Mortensen15,16,17, Bryan J. Mowry87,201, Markus M. Nöthen55,56, Roel A. Ophoff38,80,34, Michael J. Owen6,7, Aarno Palotie2,4,161, Carlos N. Pato110, Tracey L. Petryshen2,128,202, Danielle Posthuma203,204,205, Marcella Rietschel77, Brien P. Riley196, Dan Rujescu81,83, Pak C. Sham43,44,116 Pamela Sklar82,91,165, David St Clair206, Daniel R. Weinberger178,207, Jens R. Wendland166, Thomas Werge17,90,208, Mark J. Daly1,2,3, Patrick F. Sullivan26,51,160 & Michael C. O’Donovan6,7
Affiliations: 1Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, Massachusetts 02114, USA. 2Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, Massachusetts 02142, USA. 3Medical and Population Genetics Program, Broad Institute of MIT and Harvard, Cambridge, Massachusetts 02142, USA. 4Psychiatric and Neurodevelopmental Genetics Unit, Massachusetts General Hospital, Boston, Massachusetts 02114, USA. 5Neuropsychiatric Genetics Research Group, Department of Psychiatry, Trinity College Dublin, Dublin 8, Ireland. 6MRC Centre for Neuropsychiatric Genetics and Genomics, Institute of Psychological Medicine and Clinical Neurosciences, School of Medicine, Cardiff University, Cardiff, CF24 4HQ, UK. 7National Centre for Mental Health, Cardiff University, Cardiff,
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
CF24 4HQ, UK. 8Eli Lilly and Company Limited, Erl Wood Manor, Sunninghill Road, Windlesham, Surrey, GU20 6PH, UK. 9Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, King’s College London, London, SE5 8AF, UK. 10Center for Biological Sequence Analysis, Department of Systems Biology, Technical University of Denmark, DK-2800, Denmark. 11Division of Endocrinology and Center for Basic and Translational Obesity Research, Boston Children’s Hospital, Boston, Massachusetts, 02115USA. 12Department of Clinical Neuroscience, Psychiatry Section, Karolinska Institutet, SE-17176 Stockholm, Sweden. 13Department of Psychiatry, Diakonhjemmet Hospital, 0319 Oslo, Norway. 14NORMENT, KG Jebsen Centre for Psychosis Research, Institute of Clinical Medicine, University of Oslo, 0424 Oslo, Norway. 15Centre for Integrative Register-based Research, CIRRAU, Aarhus University, DK-8210 Aarhus, Denmark. 16National Centre for Register-based Research, Aarhus University, DK-8210 Aarhus, Denmark. 17The Lundbeck Foundation Initiative for Integrative Psychiatric Research, iPSYCH, Denmark. 18State Mental Hospital, 85540 Haar, Germany. 19Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford, California 94305, USA. 20Department of Psychiatry and Behavioral Sciences, Atlanta Veterans Affairs Medical Center, Atlanta, Georgia 30033, USA. 21Department of Psychiatry and Behavioral Sciences, Emory University, Atlanta Georgia 30322, USA. 22Virginia Institute for Psychiatric and Behavioral Genetics, Department of Psychiatry, Virginia Commonwealth University, Richmond, Virginia 23298, USA. 23Clinical Neuroscience, Max Planck Institute of Experimental Medicine, Göttingen 37075, Germany. 24Department of Medical Genetics, University of Pécs, Pécs H-7624, Hungary. 25Szentagothai Research Center, University of Pécs, Pécs H-7624, Hungary. 26Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm SE-17177, Sweden. 27Department of Psychiatry, University of Iowa Carver College of Medicine, Iowa City, Iowa 52242, USA. 28University Medical Center Groningen, Department of Psychiatry, University of Groningen NL-9700 RB, The Netherlands. 29School of Nursing, Louisiana State University Health Sciences Center, New Orleans, Louisiana 70112, USA. 30Athinoula A. Martinos Center, Massachusetts General Hospital, Boston, Massachusetts 02129, USA. 31Center for Brain Science, Harvard University, Cambridge, Massachusetts, 02138 USA. 32Department of Psychiatry, Massachusetts General Hospital, Boston, Massachusetts, 02114 USA. 33Department of Psychiatry, University of California at San Francisco, San Francisco, California, 94143 USA. 34University Medical Center Utrecht, Department of Psychiatry, Rudolf Magnus Institute of Neuroscience, 3584 Utrecht, The Netherlands. 35Department of Human Genetics, Icahn School of Medicine at Mount Sinai, New York, New York 10029 USA. 36Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York 10029 USA. 37Centre Hospitalier du Rouvray and INSERM U1079 Faculty of Medicine, 76301 Rouen, France. 38Department of Human Genetics, David Geffen School of Medicine, University of California, Los Angeles, California 90095, USA. 39Schizophrenia Research Institute, Sydney NSW 2010, Australia. 40School of Psychiatry, University of New South Wales, Sydney NSW 2031, Australia. 41Royal Brisbane and Women’s Hospital, University of Queensland, Brisbane, St Lucia QLD 4072, Australia. 42Institute of Psychology, Chinese Academy of Science, Beijing 100101, China. 43Department of Psychiatry, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China. 44State Key Laboratory for Brain and Cognitive Sciences, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China. 45Department of Computer Science, University of North Carolina, Chapel Hill, North Carolina 27514, USA. 46Castle Peak Hospital, Hong Kong, China. 47Institute of Mental Health, Singapore 539747, Singapore. 48Department of Psychiatry, Washington University, St. Louis, Missouri 63110, USA. 49Department of Child and Adolescent Psychiatry, Assistance Publique Hopitaux de Paris, Pierre and Marie Curie Faculty of Medicine and Institute for Intelligent Systems and Robotics, Paris, 75013, France. 50 Blue Note Biosciences, Princeton, New Jersey 08540, USA 51Department of Genetics, University of North Carolina, Chapel Hill, North Carolina 27599-7264, USA. 52Department of Psychological Medicine, Queen Mary University of London, London E1 1BB, UK. 53Molecular Psychiatry Laboratory, Division of Psychiatry, University College London, London WC1E 6JJ, UK. 54Sheba Medical Center, Tel Hashomer 52621, Israel. 55Department of Genomics, Life and Brain Center, D-53127 Bonn, Germany. 56Institute of Human Genetics, University of Bonn, D-53127 Bonn, Germany. 57Applied Molecular Genomics Unit, VIB Department of Molecular Genetics, University of Antwerp, B-2610 Antwerp, Belgium. 58Centre for Integrative Sequencing, iSEQ, Aarhus University, DK-8000 Aarhus C, Denmark. 59Department of Biomedicine, Aarhus University, DK-8000 Aarhus C, Denmark. 60First Department of Psychiatry, University of Athens Medical School, Athens 11528, Greece. 61Department of Psychiatry, University College Cork, Co. Cork, Ireland. 62Department of Medical Genetics, Oslo University Hospital, 0424 Oslo, Norway. 63Cognitive Genetics and Therapy Group, School of Psychology and Discipline of Biochemistry, National University of Ireland Galway, Co. Galway, Ireland. 64Department of Psychiatry and Behavioral Neuroscience, University of Chicago, Chicago, Illinois 60637, USA. 65Department of Psychiatry and Behavioral Sciences, NorthShore University HealthSystem, Evanston, Illinois 60201, USA. 66Department of Non-Communicable Disease Epidemiology, London School of Hygiene and Tropical Medicine, London WC1E 7HT, UK. 67Department of Child and Adolescent Psychiatry, University Clinic of Psychiatry, Skopje 1000, Republic of Macedonia. 68Department of Psychiatry, University of Regensburg, 93053 Regensburg, Germany. 69Department of General Practice, Helsinki University Central Hospital, University of Helsinki P.O. Box 20, Tukholmankatu 8 B, FI-00014, Helsinki, Finland 70Folkhälsan Research Center, Helsinki, Finland, Biomedicum Helsinki 1, Haartmaninkatu 8, FI-00290, Helsinki, Finland.71National Institute for Health and Welfare, P.O. BOX 30, FI-00271 Helsinki, Finland.72Translational Technologies and Bioinformatics, Pharma Research and Early Development, F. Hoffman-La Roche, CH-4070 Basel, Switzerland. 73Department of Psychiatry, Georgetown University School of Medicine, Washington DC 20057, USA. 74Department of Psychiatry, Keck School of Medicine of the University of Southern California, Los Angeles, California 90033, USA. 75Department of Psychiatry, Virginia Commonwealth University School of Medicine, Richmond, Virginia 23298, USA. 76Mental Health Service Line, Washington VA Medical Center, Washington DC 20422, USA. 77Department of Genetic Epidemiology in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Heidelberg , D-68159 Mannheim, Germany. 78Department of Genetics, University of Groningen, University Medical Centre Groningen, 9700 RB Groningen, The Netherlands. 79Department of Psychiatry, University of Colorado Denver, Aurora, Colorado 80045, USA. 80Center for Neurobehavioral Genetics, Semel Institute for Neuroscience and Human Behavior, University of California, Los Angeles, California 90095, USA. 81Department of Psychiatry, University of Halle, 06112 Halle, Germany. 82Division of Psychiatric Genomics, Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York 10029, USA. 83Department of Psychiatry, University of Munich, 80336, Munich, Germany.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
84Departments of Psychiatry and Human and Molecular Genetics, INSERM, Institut de Myologie, Hôpital de la Pitiè-Salpêtrière, Paris, 75013, France. 85Mental Health Research Centre, Russian Academy of Medical Sciences, 115522 Moscow, Russia. 86Neuroscience Therapeutic Area, Janssen Research and Development, Raritan, New Jersey 08869, USA. 87Queensland Brain Institute, The University of Queensland, Brisbane, Queensland, QLD 4072, Australia. 88Academic Medical Centre University of Amsterdam, Department of Psychiatry, 1105 AZ Amsterdam, The Netherlands. 89Illumina, La Jolla, California, California 92122, USA. 90Institute of Biological Psychiatry, Mental Health Centre Sct. Hans, Mental Health Services Copenhagen, DK-4000, Denmark. 91Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, New York 10029, USA. 92J. J. Peters VA Medical Center, Bronx, New York, New York 10468, USA. 93Priority Research Centre for Health Behaviour, University of Newcastle, Newcastle NSW 2308, Australia. 94School of Electrical Engineering and Computer Science, University of Newcastle, Newcastle NSW 2308, Australia. 95Division of Medical Genetics, Department of Biomedicine, University of Basel, Basel, CH-4058, Switzerland. 96Department of Genetics, Harvard Medical School, Boston, Massachusetts 02115, USA. 97Section of Neonatal Screening and Hormones, Department of Clinical Biochemistry, Immunology and Genetics, Statens Serum Institut, Copenhagen, DK-2300, Denmark. 98Department of Psychiatry, Fujita Health University School of Medicine, Toyoake, Aichi, 470-1192, Japan. 99Regional Centre for Clinical Research in Psychosis, Department of Psychiatry, Stavanger University Hospital, 4011 Stavanger, Norway. 100Rheumatology Research Group, Vall d'Hebron Research Institute, Barcelona, 08035, Spain. 101Centre for Medical Research, The University of Western Australia, Perth, WA 6009, Australia. 102The Perkins Institute for Medical Research, The University of Western Australia, Perth, WA 6009, Australia.103Department of Medical Genetics, Medical University, Sofia1431, Bulgaria. 104Department of Psychology, University of Colorado Boulder, Boulder, Colorado 80309, USA. 105Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, M5T 1R8, Canada. 106Department of Psychiatry, University of Toronto, Toronto, Ontario, M5T 1R8, Canada. 107Institute of Medical Science, University of Toronto, Toronto, Ontario, M5S 1A8, Canada. 108Institute of Molecular Genetics, Russian Academy of Sciences, Moscow123182, Russia. 109Latvian Biomedical Research and Study Centre, Riga, LV-1067, Latvia. 110Department of Psychiatry and Zilkha Neurogenetics Institute, Keck School of Medicine at University of Southern California, Los Angeles, California 90089, USA. 111Faculty of Medicine, Vilnius University, LT-01513 Vilnius, Lithuania. 112 Department of Biology and Medical Genetics, 2nd Faculty of Medicine and University Hospital Motol, 150 06 Prague, Czech Republic. 113 Department of Child and Adolescent Psychiatry, Pierre and Marie Curie Faculty of Medicine, Paris 75013, France. 114Duke-NUS Graduate Medical School, Singapore 169857, Singapore. 115Department of Psychiatry, Hadassah-Hebrew University Medical Center, Jerusalem 91120, Israel. 116Centre for Genomic Sciences, The University of Hong Kong, Hong Kong, China.117Mental Health Centre and Psychiatric Laboratory, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. 118Department of Biostatistics, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland 21205, USA. 119Department of Psychiatry, Columbia University, New York, New York 10032, USA. 120Priority Centre for Translational Neuroscience and Mental Health, University of Newcastle, Newcastle NSW 2300, Australia.121Department of Genetics and Pathology, International Hereditary Cancer Center, Pomeranian Medical University in Szczecin, 70-453 Szczecin, Poland. 122Department of Mental Health and Substance Abuse Services; National Institute for Health and Welfare, P.O. BOX 30, FI-00271 Helsinki, Finland 123Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland 21205, USA. 124Department of Psychiatry, University of Bonn, D-53127 Bonn, Germany. 125Centre National de la Recherche Scientifique, Laboratoire de Génétique Moléculaire de la Neurotransmission et des Processus Neurodégénératifs, Hôpital de la Pitié Salpêtrière, 75013, Paris, France. 126Department of Genomics Mathematics, University of Bonn, D-53127 Bonn, Germany. 127Research Unit, Sørlandet Hospital, 4604 Kristiansand, Norway. 128Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 02115, USA. 129VA Boston Health Care System, Brockton, Massachusetts 02301, USA. 130Department of Psychiatry, National University of Ireland Galway, Co. Galway, Ireland. 131Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh EH16 4SB, UK. 132Division of Psychiatry, University of Edinburgh, Edinburgh EH16 4SB, UK. 133Division of Mental Health and Addiction, Oslo University Hospital, 0424 Oslo, Norway. 134Massachusetts Mental Health Center Public Psychiatry Division of the Beth Israel Deaconess Medical Center, Boston, Massachusetts 02114, USA. 135Estonian Genome Center, University of Tartu, Tartu 50090, Estonia. 136School of Psychology, University of Newcastle, Newcastle NSW 2308, Australia. 137First Psychiatric Clinic, Medical University, Sofia 1431, Bulgaria. 138Department P, Aarhus University Hospital, DK-8240 Risskov, Denmark. 139Department of Psychiatry, Royal College of Surgeons in Ireland, Dublin 2, Ireland. 140King’s College London, London SE5 8AF, UK. 141Maastricht University Medical Centre, South Limburg Mental Health Research and Teaching Network, EURON, 6229 HX Maastricht, The Netherlands. 142Institute of Translational Medicine, University of Liverpool, Liverpool L69 3BX, UK. 143Max Planck Institute of Psychiatry, 80336 Munich, Germany. 144Munich Cluster for Systems Neurology (SyNergy), 80336 Munich, Germany. 145Department of Psychiatry and Psychotherapy, Jena University Hospital, 07743 Jena, Germany. 146Department of Psychiatry, Queensland Brain Institute and Queensland Centre for Mental Health Research, University of Queensland, Brisbane, Queensland, St Lucia QLD 4072, Australia. 147Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA. 148Department of Psychiatry, Trinity College Dublin, Dublin 2, Ireland. 149Eli Lilly and Company, Lilly Corporate Center, Indianapolis, 46285 Indiana, USA. 150Department of Clinical Sciences, Psychiatry, Umeå University, SE-901 87 Umeå, Sweden. 151DETECT Early Intervention Service for Psychosis, Blackrock, Co. Dublin, Ireland. 152Centre for Public Health, Institute of Clinical Sciences, Queen’s University Belfast, Belfast BT12 6AB, UK. 153Lawrence Berkeley National Laboratory, University of California at Berkeley, Berkeley, California 94720, USA. 154Institute of Psychiatry, King’s College London, London SE5 8AF, UK. 155A list of authors and affiliations appear in the Supplementary Information of the original article. 156Melbourne Neuropsychiatry Centre, University of Melbourne & Melbourne Health, Melbourne, Vic 3053, Australia. 157Department of Psychiatry, University of Helsinki, P.O. Box 590, FI-00029 HUS, Helsinki, Finland. 158Public Health Genomics Unit, National Institute for Health and Welfare, P.O. BOX 30, FI-00271 Helsinki, Finland 159Medical Faculty, University of Belgrade, 11000 Belgrade, Serbia. 160Department of Psychiatry, University of North Carolina, Chapel Hill, North Carolina 27599-7160, USA. 161Institute for Molecular Medicine Finland, FIMM, University of Helsinki, P.O. Box 20FI-00014, Helsinki, Finland 162Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
02115, USA. 163Department of Psychiatry, University of Oxford, Oxford, OX3 7JX, UK. 164Virginia Institute for Psychiatric and Behavioral Genetics, Virginia Commonwealth University, Richmond, Virginia 23298, USA. 165Institute for Multiscale Biology, Icahn School of Medicine at Mount Sinai, New York, New York 10029, USA. 166PharmaTherapeutics Clinical Research, Pfizer Worldwide Research and Development, Cambridge, Massachusetts 02139, USA. 167Department of Psychiatry and Psychotherapy, University of Gottingen, 37073 Göttingen, Germany. 168Psychiatry and Psychotherapy Clinic, University of Erlangen, 91054 Erlangen, Germany. 169Hunter New England Health Service, Newcastle NSW 2308, Australia. 170School of Biomedical Sciences and Pharmacy, University of Newcastle, Callaghan NSW 2308, Australia. 171Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, Maryland 20892, USA. 172University of Iceland, Landspitali, National University Hospital, 101 Reykjavik, Iceland. 173Department of Psychiatry and Drug Addiction, Tbilisi State Medical University (TSMU), N33, 0177 Tbilisi, Georgia. 174Research and Development, Bronx Veterans Affairs Medical Center, New York, New York 10468, USA. 175Wellcome Trust Centre for Human Genetics, Oxford, OX3 7BN, UK. 176deCODE Genetics, 101 Reykjavik, Iceland. 177Department of Clinical Neurology, Medical University of Vienna, 1090 Wien, Austria. 178Lieber Institute for Brain Development, Baltimore, Maryland 21205, USA. 179Department of Medical Genetics, University Medical Centre Utrecht, Universiteitsweg 100, 3584 CG, Utrecht, The Netherlands. 180Berkshire Healthcare NHS Foundation Trust, Bracknell RG12 1BQ, UK. 181Section of Psychiatry, University of Verona, 37134 Verona, Italy. 182Department of Psychiatry, University of Oulu, P.O. BOX 5000, 90014, Finland183University Hospital of Oulu, P.O.BOX 20, 90029 OYS, Finland.184Molecular and Cellular Therapeutics, Royal College of Surgeons in Ireland, Dublin 2, Ireland. 185Health Research Board, Dublin 2, Ireland. 186School of Psychiatry and Clinical Neurosciences, The University of Western Australia, Perth WA6009, Australia. 187Computational Sciences CoE, Pfizer Worldwide Research and Development, Cambridge, Massachusetts 02139, USA.188Human Genetics, Genome Institute of Singapore, A*STAR, Singapore 138672, Singapore. 189A list of authors and affiliations appear in the Supplementary Information in the original article.190University College London, London WC1E 6BT, UK. 191Department of Neuroscience, Icahn School of Medicine at Mount Sinai, New York, New York 10029, USA. 192Institute of Neuroscience and Medicine (INM-1), Research Center Juelich, 52428 Juelich, Germany. 193Department of Genetics, The Hebrew University of Jerusalem, 91905 Jerusalem, Israel. 194Neuroscience Discovery and Translational Area, Pharma Research and Early Development, F. Hoffman-La Roche, CH-4070 Basel, Switzerland. 195Centre for Clinical Research in Neuropsychiatry, School of Psychiatry and Clinical Neurosciences, The University of Western Australia, Medical Research Foundation Building, Perth WA 6000, Australia. 196Virginia Institute for Psychiatric and Behavioral Genetics, Departments of Psychiatry and Human and Molecular Genetics, Virginia Commonwealth University, Richmond, Virginia 23298, USA. 197The Feinstein Institute for Medical Research, Manhasset, New York, 11030 USA. 198The Hofstra NS-LIJ School of Medicine, Hempstead, New York, 11549 USA. 199The Zucker Hillside Hospital, Glen Oaks, New York,11004 USA. 200Saw Swee Hock School of Public Health, National University of Singapore, Singapore 117597, Singapore. 201Queensland Centre for Mental Health Research, University of Queensland, Brisbane 4076, Queensland, Australia. 202Center for Human Genetic Research and Department of Psychiatry, Massachusetts General Hospital, Boston, Massachusetts 02114, USA. 203Department of Child and Adolescent Psychiatry, Erasmus University Medical Centre, Rotterdam 3000, The Netherlands. 204Department of Complex Trait Genetics, Neuroscience Campus Amsterdam, VU University Medical Center Amsterdam, Amsterdam 1081, The Netherlands. 205Department of Functional Genomics, Center for Neurogenomics and Cognitive Research, Neuroscience Campus Amsterdam, VU University, Amsterdam 1081, The Netherlands. 206University of Aberdeen, Institute of Medical Sciences, Aberdeen, AB25 2ZD, UK. 207Departments of Psychiatry, Neurology, Neuroscience and Institute of Genetic Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland 21205, USA. 208Department of Clinical Medicine, University of Copenhagen, Copenhagen 2200, Denmark.209Departments of Psychiatry and Human Genetics, University of Chicago, Chicago, Illinois 60637, USA.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Overview of methods for establishing the targetome of a miRNA It is widely accepted that miRNA bind to the 3’-UTR, but there is also evidence for binding in coding regions1,2. This pairing involves in most cases a so-called seed sequence with certain patterns of complementarity. However, it has also been shown that miRNA interact without such seed sequences albeit with lower average effects3. In order to establish a list of all genes, that are targeted by a given miRNA (the so-called targetome of a miRNA), computational prediction have been used extensively. In these computational prediction methods, an array of different characteristics are taken into account such as conservation across species, local AU content, predicted accessibility due to RNA folding, Gibbs energy of the interaction, and expression data. However, the prediction algorithms still suffer from limited specificity and/or sensitivity despite over a decade of research. As an alternative to these computational prediction methods various experimental methods exist. These include reporter assays, which suffer from their low throughput, and overexpression/silencing experiments measuring the change in protein abundance resulting from altered miRNA expression. Another experimental approach is cross-linked immunoprecipitation (CLIP) experiments where the miRNA, its associated proteins, and the fragment of mRNA that it binds are precipitated and subsequently RNA sequencing is carried out. This gives the abundance of each miRNA expressed in the tissue/cell as well as the target sites on the genes. However, subsequently computer algorithms have to be employed to predict which miRNA binds to which target sites. A relatively new approach which circumvents this limitation is “CLASH” which is similar to CLIP experiments, but with the difference that the miRNA and the fragment of the mRNA to which it binds are ligated3. Unfortunately, this has only been used in yeast and in HEK cells and the authors demonstrated a depletion of brain-expressed miRNA in the experiments. In our paper, we opted to primarily use the predicted conserved predicted targets of the widely used TargetScan. Because of its reliance on conservation and that the miRNA have a seed site, the predicted targets of this algorithm have a relatively high chance of being both true predictions and functionally important. For validation purposes, two alternative prediction methods, were used: The recent methods TargetMiner4, which is based on mirBase 18, and the broadly used but slightly older MiRanda5, which is based on mirBase 15, were used.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Filtering of predicted miRNA targets with CLIP data We used data from AGO cross-linked immunoprecipitation (CLIP) experiments in an attempt to remove false predictions from our TargetScan gene sets. Results from 41 CLIP experiments in human and mouse were down-loaded from StarBase6. Additionally, one experiment from human brain7 and a group of experiments including mainly hematopoietic cells8 were included. LiftOver was used to lift results from hg18 and from mm9 to hg19. Peaks longer than 300 bp were excluded from the analysis due to lack of specificity (this represented 0.6% of all peaks). A miRNA target site was defined to be supported by CLIP data if the seed-region of the target site was completely within the peak. In a subsequent analysis we determined whether filtering of our top ranking conserved miRNA gene sets with the aforementioned CLIP data led to an increased enrichment for schizophrenia associated genes. In other words, we tested whether filtering with CLIP data was better than removing genes at random. For each of our top miR-NA gene sets, 1000 replicate gene sets were generated by drawing genes from the original non-filtered gene set until the replicate set contained the same number of genes as the CLIP filtered gene set. Subsequently, the IN-RICH p-values of these replicate sets under the top-1% clumping threshold were compared to the p-values of the original CLIP filtered gene sets by counting how many replicate sets that had as strong or a stronger association to schizophrenia. This resulted in empirical p-values for the following four top ranked miRNA: miR-9-5p ~ p=0.16; miR-485-5p ~ p=0.064; miR-137 ~ p=0.16; and miR-101-3p ~ p=0.24. While filtering with CLIP data showed an overall trend to improve our enrichment p-values there was no significant evidence that filtering with CLIP data was better than removing genes at random. We note that our analysis was limited by the availability of relevant CLIP data and that only two brain related datasets could be incorporated. Thus, many of the CLIP supported targets in our top gene sets therefore come from samples other than the two CLIP experiments from brain. For miR-9-5p, the human brain experiment has 6 targets (0.49% of targets from that experiment) and the p13 mouse brain contains 143 (3.3% of targets from that experiment).
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
The correlation between the number of miRNA binding sites in a gene and schizophrenia To determine whether miRNA regulated genes were more likely to be schizophrenia risk genes, we tested if genes under extensive miRNA regulation showed a stronger association with schizophrenia compared to genes with reduced or missing miRNA regulation. More precisely, we tested if there was a correlation between the number of predicted miRNA target sites per gene and the corresponding strength of association with schizophre-nia. The p-value assigned to a gene (strength of association) was the p-value of the most significantly associated SNP physically located within the gene (min p). Genes in the broader MHC-region were excluded. A linear model for the gene p-value was run including the following independent variables: the number of miRNA target sites in the gene, the log-transformed gene length, and the 3’-UTR length. This model looked like this:
gene length ‐3.165e‐06 8.009e‐08 ‐39.522 <2e‐16 This shows that our covariates are also associated with the gene p-value. We further note that miRNA target sites are also significantly correlated with the 3’UTR length and the gene length and that the identified association was in the same order of magnitude when omitting the correction for 3’-UTR length (data not shown). We subsequently calculated the fraction of the variance in the GWAS results solely explained by the number of miRNA target sites, regressing out effects of gene 3’-UTR length. In addition to the calculations for schizophre-nia both of the two abovementioned approaches were also repeated for the three unrelated traits9-12 . The results for the correlation between miRNA gene target sites and gene p-value were (sample size is number of cases and controls in the respective study):
Trait beta p‐value variance explained Sample size
Schizophrenia9 ‐1.59E‐02 <2e‐16 0.49% 150,064
Age at Menarche10 ‐8.20E‐03 3.00E‐09 0.20% 182,416
Crohn's11 ‐3.69E‐03 0.06169 0.02% 66,968
Height12 ‐5.41E‐02 <2e‐16 0.52% 253,288
For schizophrenia the analysis shows a trait unspecific correlation of gene p-values with the number of miRNA target sites per gene. In other words, genes with more miRNA target sites on average had a lower p-value. Simi-lar associations were found for age at menarche and height but not for Crohn’s disease.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Circos plot for the MHC region (targets of top-10 miRNA) To provide an overview of the targets of the top-10 miRNA in the MHC-region, we provide eFigure 1. eFigure 1: Circos plot for the MHC-region.
A zoomed in version of the MHC-region of the Circos plot in the main article. Note that the genes in this region were excluded from the gene set analyses. GW: genome-wide.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Clustering of top 10 miRNA gene sets considering only schizophrenia associ-ated genes To see how the top-10 scoring miRNA gene sets related when considering only schizophrenia associated genes, they were clustered considering only genes at various cut-offs for the gene’s association with schizophrenia (eFigure 2). Genes in the broader MHC-region and on sex-chromosomes were excluded in the analysis. eFigure 2: Clustering of the top-10 scoring miRNA families.
The clustering was based on the Jaccard distance between their target genes on autosomes outside the MHC-region considering a) all genes, b) genes within top 5% of SNPs c) genes within top 1% of SNPs d) genes with p < 1 x 10-5, and e) genes with p < 5 x 10-8 (Genome-wide significant).
miR−
7−5p
miR−
34a
c−5
p/44
9b−
5p/4
49a
miR−
1/20
6/61
3
miR−
374
ab−
5p
miR−
200
bc−
3p/4
29
miR−
101−
3p miR−
9−
5p
miR−
137
miR−
28−
5p/7
08−
5p/3
139
miR−
485−
5p
0.8
80.
900.
920
.94
0.9
6
miR−
7−5
p
miR−
34a
c−5p
/449
b−
5p/4
49a
miR−
1/20
6/6
13
miR−
374a
b−5
p
miR−
9−5p
miR−
137
miR−
200
bc−
3p/4
29
miR−
101−
3p
miR−
28−
5p/7
08−
5p/3
139
miR−
485−
5p
0.8
60.
880
.90
0.9
20.
940.
96
miR−
28−
5p/7
08−
5p/3
139
miR−
485−
5p
miR−
7−5p
miR−
34ac−
5p/4
49b−
5p/4
49a
miR−
1/20
6/61
3
miR−
374
ab−
5p
miR−
137
miR−
9−5
p
miR−
200
bc−
3p/4
29
miR−
101−
3p0.8
20
.84
0.86
0.88
0.90
0.9
20
.94
0.96
miR−
7−5
p
miR−
374a
b−5p
miR−
101−
3p
miR−
200b
c−3p
/42
9
miR−
9−5p
miR−
137
miR−
1/20
6/6
13
miR−
34a
c−5p
/449
b−
5p/4
49a m
iR−
28−
5p/7
08−
5p/3
139
miR−
485−
5p
0.8
00.
850.
900.
95
miR−
485−
5p
miR−
28−
5p/7
08−
5p/3
139
miR−
9−5p
miR−
7−5p
miR−
200b
c−3p
/42
9
miR−
374a
b−5p
miR−
101−
3p
miR−
1/20
6/6
13
miR−
137
miR−
34a
c−5p
/44
9b−
5p/4
49a
0.6
50.
700.
750.
800.
850
.90
0.95
1.00
a) b) c)
d) e)
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Regional association plots of the three mir-9 host genes Ricopili (http://www.broadinstitute.org/mpg/ricopili/) was used to visualize association results from PGC2 in genomic regions harboring variants related to mir-9 (eFigure 3). eFigure 3: Region plots of GWAS associations for the miR-9-5p host genes.
The “PGC2” associations at a) MIR9-1 b) MIR9-2 c) MIR9-3. LINC00461 is the putative host gene of MIR9-2. The lead SNP (rs181900) is 25kb upstream of the longest isoform of this host gene and 43kb upstream of MIR9-2 itself.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Temporal brain expression of miR-9-5p, miR-137, and miR-2682-5p To establish the temporal expression pattern of selected miRNA, BrainSpan RNA sequencing data was used13. BrainSpan data includes a variable number of samples at various regions at various time points. In our analyses, only regions present in 30 or more samples were used. When more than one sample at a given region and devel-opmental stage was present, the average expression was used. The results of this analysis is presented in eFigure 4. eFigure 4: Spatiotemporal brain expression of MIR137, MIR2682, and MIR9-2.
MIR9-2 is the mir-9 expressing gene showing the highest expression in BrainSpan, and the second peak at 26 weeks in STC seen in its figure is from a single sample at a single time point, and can thus be an outlier. A1C: primary auditory cortex (core), AMY: amygdaloid complex, DFC: dorsolateral pre-frontal cortex, HIP: hippocampus (hippocampal formation), IPC: posteroventral (inferior) parietal cor-tex, ITC: inferolateral temporal cortex (area TEv), MFC: anterior (rostral) cingulate (medial prefrontal) cortex, OFC: orbital frontal cortex, STC: posterior (caudal) superior temporal cortex (area 22c), URL: upper (rostral) rhombic lip, V1C: primary visual cortex (striate cortex), VFC: ventrolateral prefrontal cortex.
01
23
4
100 1000 10000Age in days since conception
Ave
rag
e R
PK
M
a) MIR137
05
101
5
100 1000 10000Age in days since conception
Ave
rag
e R
PK
M
b) MIR2682
03
060
90
120
100 1000 10000Age in days since conception
Ave
rag
e R
PK
M
c) MIR9
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Clustering of miR-9-5p targets In an attempt to identify homogeneous subsets (functional groups) in the genes targeted by miR-9-5p, we em-ployed a hierarchically clustering approach using data from BrainSpan13. Brain RNA sequencing reads per kilobase per million mapped tags (RPKM) were averaged across different developmental time points and brain regions (see below) and log transformed with log2(1+RPKM) following guiding principles from BrainSpan. All miRNA targets in a gene set were hierarchically clustered using a Pearson correlation as similarity measure. We used data from all available brain regions at developmental stages prior to and including two years of age. The age cut-off was selected in accordance with the expression pattern of miR-9-5p (eFigure 4). Genes with no ex-pression data were excluded from the analyses. A cut-off height of 0.25 resulted in four groups of targets which had a gene set size larger than or equal to 50. To test a potential association of the resulting gene sets with schiz-ophrenia we used INRICH at the top-1% threshold:
One of the gene sets, cluster 4 (eTable 15), was significantly associated with schizophrenia. While the enrich-ment for cluster 4 was less significant when compared to the enrichment for the full set, the difference was only marginal. Since this cluster included less then half of the original targets we tested in a resampling procedure whether this group of genes was enriched for genes associated with schizophrenia. The INRICH p-value of this cluster was compared to that of 1000 size-matched gene sets drawn at random from all miR-9-5p targets in the original test. Only four of these randomly drawn subsets showed as strong or a stronger p-value for association in INRICH compared to the true subset. This resulted in an empirical p-value for the change in enrichment of p = (4+1)/(1000+1) = 5.0 x 10-3. Hence, taken the overall number of genes in this subgroup into account the cluster of 497 genes in the subgroup are relatively more enriched for schizophrenia associated genes than the original set even though the subset itself had a slightly less significant p-value than the original set..
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
PPI network for miR-9-5p targets eFigure 5: PPI-network of the miR-9-5p targetome. Hierarchical clustering using data from BrainSpan (see above) is one way to identify homogeneous subsets (functional groups) in genes targeted by miR-9-5p. Another way is to use data for protein-protein interactions (PPIs). We therefore used data from the TissueNet Brain PPIs14 to see if the subset of 497 genes (cluster 4), that was identified using BrainSpan data, had more PPIs than a random subset. For this we compared the number of PPIs in the original set to size-matched gene sets drawn randomly from all miR-9-5p targets. Of 100,000 gene sets two had as many or more PPIs as that of the original gene set yielding an empirical p-value of 3.0 x 10-5. We would like to note that co-expressed genes are in general expected to be enriched for PPIs and our finding does not itself provide further evidence for the involvement of the subset in schizophrenia. Instead, it adds credibility to our clustering approach as with a different data-set we are able to show that the genes have likely more shared functionality than a random subset. The PPI-network derived from the aforementioned TissueNet data for the original set of miR-9-5p targets and the subset of genes in cluster 4 is illustrated in eFigure 5.
Members of the identified subset and interactions between its members are colored in red. Genes with no PPIs are not shown.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Analysis of the miR-9-5p targetome in post mortem brains The information from BrainSpan and TissueNet (see above) are disease unspecific and based on information from healthy individuals. In order to further characterize the miR-9-5p targetome in relation to schizophrenia, we therefore looked at the enrichment of the miR-9-5p targetome in co-expression modules from post mortem brains of patients with schizophrenia and controls. Details on how these co-expression modules have been de-rived, the characterization of the modules and a full list of their genes can be found elsewhere15. The subsequent table shows p-values for the enrichment of genes in the aforementioned co-expression modules in both the full set of targets and the subset of 497 targets. Only nominally significant modules are listed. None of the modules survives correction for multiple testing (data not shown). Additionally, it is indicated whether these modules were differentially expressed in patients compared to controls. DEiS – Module differentially expressed in schiz-ophrenia (yes/no); # Genes – Number of targets in module.
Module p‐value for enrichment in module DEiS # Genes
All miR‐9‐5p targets (1237 genes)
MG3 2.2E‐04 yes 16
PV(+)1 7.6E‐04 no 125
OLIG3 4.1E‐03 no 13
UNK5 1.7E‐02 no 9
OLIG1 3.7E‐02 Yes 107
miR‐9‐5p subset (497 genes)
UNK1 3.0E‐04 no 44
UNK5 2.8E‐02 no 5
OLIG3 3.5E‐02 no 6
MG3 4.1E‐02 yes 6
Genes in modules Genes located in GWAS regions = clumps with p‐values < 1 x 10‐5 in PGC2 are underlined and printed in bold if the p‐value was < 5 x 10‐8 in PGC2. All miR‐9‐5p targets: MG3: ADAMTS9, ANXA2, CCDC50, CHSY1, CMTM6, COL4A2, EHD4, IL6R, ITGA6, KCTD12, MYH9, NRP1, TFRC, TGFBR1, TGFBR2, TMEM109
We would like to note that due to a limited size of our sample under study we were a priori less likely to detect a true schizophrenia associated sub-module. This is unfortunately true for most currently available post-mortem data-sets in schizophrenia. New, yet to be published datasets will provide more power to unequivocally detect , schizophrenia associated sub-modules and we keep confident that these analyses will add further support to our finding. Furthermore we would like to note that post mortem expression analyses might not capture all of the differential expression relating to the neurodevelopmental role of miR-9-5p.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
Determining if the targetome overlap between two miRNA is bigger than ex-pected by chance In the main text, we look at the intersections of the predicted targets for two pairs of miRNA – miR-9-5p com-pared to miR-137 and miR-2682-5p to miR-137 (eFigure 6) eFigure 6: Venn diagrams of the miR-9-5p/miR-137 and miR-137/-miR-2682-5p targetomes.
While analyzing the overlap in these two miRNA pairs we observed an unequal distribution in the number of miRNA targeting individual genes. We observed a tendency of genes to be targeted by either many miRNA or only one/a few. As a consequence genes targeted by multiple miRNA are more likely to appear in an overlap of the targetomes of two random miRNA. To establish if an overlap between the targetomes of two miRNA is big-ger than expected by chance, it therefore is indicated to take this observation into account. Ignoring this observa-tion and taking Fisher’s exact test (in case of small cell counts) or a chi-square test would likely lead to errone-ous associations. As an alternative, the following procedure was employed (detailed for the example overlap of miR-9-5p and miR-137):
A pool of genes comprising all conserved targets of all conserved miRNA-families was created. We did so add-ing for every occurrence of a target in a miRNA targetome one copy of this gene to the combined pool. As a consequence each gene was n times present in the pool with n being the number of miRNA-families targeting this gene. Please note that no copies were added to the pool for the miR-137 targetome. In a subsequent step we created r = 1000 replicate sets of miR-9-5p targetomes matching the original number of miR-9-5p targets (n=1237) as follows: for each replicate set we drew genes at random from the pool and added them to the repli-cate set if that gene wasn’t already in the replicate set. When the replicate set had the same number of (unique) genes as the original set we calculated the overlap with miR-137. For this we counted all instances p where the replicate sets had as large or a larger overlap with miR-137 as the original targetome of miR-9-5p and calculated the empirical p-value as the fraction p/r. Note that our procedure can be interpreted as weighed sampling.
We used the same principles to study the overlap between miR-137 and miR-2682-5p. However, for this analy-sis we used all TargetScan predictions regardless of conservation (see main text for details).
The results for the overlaps with miR-137 were: Equal or larger
overlap Less overlap Empirical
p‐value Chi‐square test p‐
value
miR‐9‐5p 277 723 0.28 < 2.2 x 10‐16
miR‐2682‐5p 310 690 0.31 < 2.2 x 10‐16
The vast incongruences between the empirical p-values and the chi-square test results are likely caused by the aforementioned unequal distribution in the number miRNA targeting a given gene. On average a random gene targeted by at least one miRNA are targeted by 6.5 miRNA, whereas the genes in the intersection of miR-137 and miR-9-5p on average are targeted by 20 miRNA. The fact that the two overlaps are non-significant does not preclude shared functions between the miRNA, but it indicates that further biological knowledge would be need-ed to establish it. It should be noted that we choose to study the overlap in targetomes of miR-9-5p and miR-137 based on our clustering results for these two miRNA targetomes (as part of the ten highest scoring miRNA gene sets, see Figure 2b and eFigure 2). It should also noted that the miR-2682-5p targetome likely has more false predictions due to the inclusion of non-conserved targets.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
A note on the limited support for an association of the miR-130b and miR-193a-3p targetomes As detailed in the main text, imprecision in target prediction is a limitation to our study. This might be one of the reasons why our study only found modest evidence for association with miR-130b, a miRNA previously impli-cated with schizophrenia16. Another reason might the sample size of the PGC2 study. While sufficiently powered to detect association with 108 schizophrenia loci it still lacks power to detect all disease-associated variants at reasonable significance levels17. Both reasons could potentially also underlie the complete lack of evidence for association for miR-193a-3p, the second miRNA implicated in the aforementioned study16.
The three different thresholds represent the different significance thresholds for the index-SNP used during the clumping procedures. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. For a detailed description of the “Score” column please refer to the main text. “Size” indicates the number of genes in the gene set. “Brain” indicates the percentage of genes in the targetomes expressed in the brain. ”Corrected” are the p-values after correcting for multiple testing of all conserved miRNA families, where-as “Raw” are the unadjusted p-values. miRNAs are ranked by their Score.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
eReferences 1. Duursma AM, Kedde M, Schrier M, le Sage C, Agami R. miR-148 targets human DNMT3b
protein coding region. Rna. 2008;14(5):872-877. 2. Forman JJ, Legesse-Miller A, Coller HA. A search for conserved sequences in coding regions
reveals that the let-7 microRNA targets Dicer within its coding sequence. Proceedings of the National Academy of Sciences of the United States of America. 2008;105(39):14879-14884.
3. Helwak A, Kudla G, Dudnakova T, Tollervey D. Mapping the human miRNA interactome by CLASH reveals frequent noncanonical binding. Cell. 2013;153(3):654-665.
4. Bandyopadhyay S, Mitra R. TargetMiner: microRNA target prediction with systematic identification of tissue-specific negative examples. Bioinformatics. 2009;25(20):2625-2631.
5. Betel D, Koppal A, Agius P, Sander C, Leslie C. Comprehensive modeling of microRNA targets predicts functional non-conserved and non-canonical sites. Genome biology. 2010;11(8):R90.
6. Li JH, Liu S, Zhou H, Qu LH, Yang JH. starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. Nucleic acids research. 2014;42(Database issue):D92-97.
7. Boudreau RL, Jiang P, Gilmore BL, et al. Transcriptome-wide discovery of microRNA binding sites in human brain. Neuron. 2014;81(2):294-305.
8. Balakrishnan I, Yang X, Brown J, et al. Genome‐Wide Analysis of miRNA‐mRNA Interactions in Marrow Stromal Cells. Stem Cells. 2014;32(3):662-673.
9. Schizophrenia Working Group of the Psychiatric Genomics Consortium. Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421-427.
10. Perry JR, Day F, Elks CE, et al. Parent-of-origin-specific allelic associations among 106 genomic loci for age at menarche. Nature. 2014;514(7520):92-97.
11. Jostins L, Ripke S, Weersma RK, et al. Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease. Nature. 2012;491(7422):119-124.
12. Wood AR, Esko T, Yang J, et al. Defining the role of common variation in the genomic and biological architecture of adult human height. Nature genetics. 2014;46(11):1173-1186.
13. BrainSpan: Atlas of the Developing Human Brain. 2011; Funded by ARRA Awards 1RC2MH089921-089901, 089921RC089922MH090047-089901, and 089921RC089922MH089929-089901. Available at: http://brainspan.org. Accessed 17 Jun, 2014.
14. Barshir R, Basha O, Eluk A, Smoly IY, Lan A, Yeger-Lotem E. The TissueNet database of human tissue protein–protein interactions. Nucleic acids research. 2013;41(D1):D841-D844.
15. Roussos P, Katsel P, Davis KL, Siever LJ, Haroutunian V. A system-level transcriptomic analysis of schizophrenia using postmortem brain tissue samples. Archives of general psychiatry. 2012;69(12):1205-1213.
16. Wei H, Yuan Y, Liu S, et al. Detection of Circulating miRNA Levels in Schizophrenia. American Journal of Psychiatry. 2015:appi. ajp. 2015.14030273.
17. Ripke S, O'Dushlaine C, Chambert K, et al. Genome-wide association analysis identifies 13 new risk loci for schizophrenia. Nature genetics. 2013;45(10):1150-1159.
Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/25/2021
eTable 1: Full results of the TargetScan conserved miRNA gene set analysis. The three different thresholds represent the different significance thresholds for the index-SNP used during the clumping procedures. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. For a detailed description of the “Score” column please refer to the main text. “Size” indicates the number of genes in the gene set. “Brain” indicates the per-centage of genes in the targetomes expressed in the brain. ”Corrected” are the p-values after correct-ing for multiple testing, whereas “Raw” are the unadjusted p-values. miRNAs are ranked by their Score. miRNA with less than 50 targets were not included in our tests and therefore the p-values are marked with “NA” instead.
eTable 2: Alternative TargetScan gene set analysis: longer clumping range. These are the results of the main gene set analysis (eTable 1) with the change that the clumping range was increased to 3000 kb.
eTable 3: Alternative TargetScan gene set analysis: lower LD-threshold. These are the results of the main gene set analysis (eTable 1) with the change that the clumping LD threshold was lowered to 0.1.
eTable 4: Alternative TargetScan gene set analysis: lower LD-threshold and longer clumping range. These are the results of the main gene set analysis (eTable 1) with the change that the clumping range was 3000 kb and the LD threshold was lowered to 0.1.
eTable 5: Validation of top-10 gene sets using MiRanda target predictions. Of the two groups of predictions in MiRanda, only the group “Good mirSVR score” was used. In order to match these algorithms to the TargetScan miRNA families, the union of predictions for the individual miRNA members in the TargetScan family was taken. The three different thresholds represent the different significance thresholds for the index-SNP used in clumping. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. “OSF = Original set found” indicates the fraction of the TargetScan predictions also predicted by MiRanda. ”Corrected” are the p-values after correcting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text. MiRanda top‐10 validation (R2=0.6, 500kb)
eTable 6: Validation of top-10 gene sets using TargetMiner target predictions. In order to match these algorithms to the TargetScan miRNA families, the union of predictions for the individual miRNA members in the TargetScan family was taken. The three different thresholds represent the different significance thresholds for the index-SNP used in clumping. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. “OSF = Original set found” indicates the fraction of the TargetScan predictions also predicted by MiRanda. ”Corrected” are the p-values after correcting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text. TargetMiner top‐10 validation (R2=0.6, 500kb)
eTable 7: Results for the top-10 gene sets filtered with CLIP-data from 58 ex-periments. miRNA target sites were defined to be supported by CLIP data if the seed-region of the target sites were completely within a CLIP peak. The three different thresholds represent the different significance thresholds for the index-SNP used in clumping. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. “OSF = Original set found” indicates the fraction of the TargetScan predictions also predicted by MiRanda. ”Corrected” are the p-values after correcting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text. TargetScan top‐10 CLIP‐filtered (R2=0.6, 500kb)
eTable 8: Signal for the top-10 gene sets in three unrelated traits. The top 10 conserved miRNA gene sets (Table 1), were tested in three well powered alternative GWAS traits10-12 to assess if the signal was specific or a general finding common to many traits. The three different thresholds represent the different significance thresholds for the index-SNP used in clumping. “Size” indicates the number of genes in the gene set. “Size” indicates the number of genes in the gene set. “Brain” indicates the percentage of the test genes expressed in the brain. ”Corrected” are the p-values after correcting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text.
eTable 9: Functional annotation of genes regulated by the top-10 miRNA – bio-logical processes. The following table contains a DAVID analysis showing enrichment of GO biological processes for genes targeted by two or more of the top ten conserved miRNA (Table 1). Only the hundred most significant processes are shown. “FE” is fold enrichment. The list of all genes in the categories can be obtained from the authors by request.
Top 10 miRNA GO Biological Process Count % P‐value List Total
Pop Hits
Pop Total FE
Benja‐mini
regulation of transcription 388 22 2.2E‐22 1303 2601 13528 1.5 7.4E‐19regulation of transcription from RNA polymerase II promoter 150 9 4.0E‐20 1303 727 13528 2.1 6.6E‐17positive regulation of transcription 123 7 1.5E‐18 1303 564 13528 2.3 1.7E‐15positive regulation of transcription, DNA‐dependent 110 6 1.7E‐18 1303 477 13528 2.4 1.4E‐15positive regulation of RNA metabolic process 110 6 3.3E‐18 1303 481 13528 2.4 2.2E‐15positive regulation of gene expression 124 7 6.7E‐18 1303 581 13528 2.2 3.7E‐15positive regulation of nucleobase, nucleoside, nucleotide and nu‐cleic acid metabolic process
130 8 7.6E‐18 1303 624 13528 2.2 3.6E‐15
positive regulation of macromolecule biosynthetic process 133 8 2.4E‐17 1303 654 13528 2.1 1.0E‐14positive regulation of nitrogen compound metabolic process 130 8 2.2E‐16 1303 644 13528 2.1 8.2E‐14positive regulation of biosynthetic process 136 8 4.9E‐16 1303 695 13528 2.0 1.5E‐13positive regulation of cellular biosynthetic process 134 8 8.0E‐16 1303 685 13528 2.0 2.3E‐13transcription 307 18 1.4E‐15 1303 2101 13528 1.5 4.0E‐13negative regulation of gene expression 108 6 1.4E‐15 1303 504 13528 2.2 3.7E‐13negative regulation of transcription 101 6 2.1E‐15 1303 459 13528 2.3 5.0E‐13regulation of RNA metabolic process 270 16 1.4E‐14 1303 1813 13528 1.5 3.0E‐12positive regulation of transcription from RNA polymerase II pro‐moter
85 5 4.2E‐14 1303 371 13528 2.4 8.7E‐12
positive regulation of macromolecule metabolic process 152 9 4.5E‐14 1303 857 13528 1.8 8.8E‐12negative regulation of macromolecule biosynthetic process 110 6 6.7E‐14 1303 547 13528 2.1 1.2E‐11regulation of transcription, DNA‐dependent 260 15 3.2E‐13 1303 1773 13528 1.5 5.5E‐11negative regulation of cellular biosynthetic process 110 6 3.8E‐13 1303 561 13528 2.0 6.3E‐11negative regulation of biosynthetic process 111 6 6.8E‐13 1303 573 13528 2.0 1.1E‐10protein amino acid phosphorylation 123 7 1.2E‐12 1303 667 13528 1.9 1.8E‐10negative regulation of transcription, DNA‐dependent 79 5 2.1E‐12 1303 356 13528 2.3 3.0E‐10negative regulation of nucleobase, nucleoside, nucleotide and nucleic acid metabolic process
eTable 10: Functional annotation of genes regulated by the top-10 miRNA – annotation clusters. The following table contains a DAVID analysis showing enrichment of functional clusters for genes targeted by two or more of the top ten conserved miRNA (Table 1). Only the ten most significant annotation clusters are shown. “FE” is fold enrichment. The list of all genes in the categories can be obtained from the authors by request. Top 10 miRNA Annotation clusters Score/Term Count % P‐value
eTable 11: The top 50 most schizophrenia-associated genes targeted by miR-9-5p. ”p-value” is the p-value of association of the most strongly associated variant located within the boundaries of the gene, and ”SNP” is the name of this variant. Note that these p-values are without taking the PGC2 replication into account and do not refer to the clump definition from PGC2 (see above). The list therefore might differ with regards to the published p-values from PGC2 for these genes. Entrez Gene p‐value SNP Description 54805 CNNM2 3.64E‐16 rs7913682 cyclin M2
23368 PPP1R13B 5.61E‐13 rs4906364 protein phosphatase 1, regulatory subunit 13B
eTable 12: Functional annotation of genes regulated by miR-9-5p – biological processes. The following table contains a DAVID analysis showing enrichment of GO biological processes for genes targeted by miR-9-5p. Only the hundred most significant processes are shown. “FE” is fold enrichment. The list of all genes in the categories can be obtained from the authors by request.
miR‐9‐5p DAVID GO Biological Process annotation Count % P‐value List Total
eTable 13: Functional annotation of genes regulated by miR-9-5p – annotation clusters. The following table contains a DAVID analysis showing enrichment of functional clusters for genes targeted miR-9-5p. Only the ten most significant annotation clusters are shown. “FE” is fold enrichment. The list of all genes in the categories can be obtained from the authors by request. miR‐9‐5p Annotation clusters Term Count % P‐value
eTable 14: Functional annotation of genes regulated by miR-9-5p – transcrip-tion factors. The following table contains a DAVID analysis showing enrichment of transcription factor regulation for genes targeted by miR-9-5p. Only the hundred most significant transcription factors are shown. “FE” is fold enrichment. The list of all genes in the categories can be obtained from the authors by request.
miR‐9‐5p DAVID Transcription Factor regulation Count % P‐value List Total
eTable 15: Genes in the schizophrenia enriched miR-9-5p subset. This subset of 497 co-expressed gene targets of the miR-9-5p targetome was found enriched in schizophrenia genes. Genes located in GWAS regions (clumps) with p-values < 1 x 10-5 in PGC2 are underlined and additionally printed in bold if the p-value was < 5 x 10-8 in PGC2. Note that these p-values are taking the PGC2 replication into account. Subset of the miR‐9‐5‐p targetome (497 genes)
eTable 16: Gene set results for miRNA in schizophrenia GWAS loci. The gene set analysis for the targets of miRNA located in genome-wide significant schizophrenia GWAS loci was performed using three different prediction methods. The number of miRNA varies between the prediction methods as not all miRNA have predictions for each method. The three different thresholds represent the differ-ent significance thresholds for the index-SNP used in clumping. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. “Size” indicates the number of genes in the gene set. “Brain” indicates the percentage of the test genes expressed in the brain. “Corrected” are the p-values after cor-recting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text.
eTable 17: Gene set results for miRNA in schizophrenia CNVs. The gene set analysis for the targets of miRNA located in schizophrenia associated CNVs was performed using three different prediction methods. The number of miRNA varies between the prediction methods as not all miRNA have predictions for each method. The three different thresholds represent the different significance thresholds for the index-SNP used in clumping. Top-1% of SNPs have p-values less than 3.420 x 10-4, top 5% of SNPs have p-values less than 1.096 x 10-2. “Size” indicates the number of genes in the gene set. “Size” indicates the number of genes in the gene set. “Brain” indicates the percentage of the test genes expressed in the brain. “Corrected” are the p-values after correcting for multiple testing, whereas “Raw” are the unadjusted p-values. For a detailed description of the “Score” column please refer to the main text.