Your browser doesn't support javascript.
loading
mixWAS: An efficient distributed algorithm for mixed-outcomes genome-wide association studies.
Li, Ruowang; Benz, Luke; Duan, Rui; Denny, Joshua C; Hakonarson, Hakon; Mosley, Jonathan D; Smoller, Jordan W; Wei, Wei-Qi; Ritchie, Marylyn D; Moore, Jason H; Chen, Yong.
Afiliación
  • Li R; Department of Computational Biomedicine, Cedars-Sinai Medical Center.
  • Benz L; Department of Biostatistics, Harvard T.H. Chan School of Public Health.
  • Duan R; Department of Biostatistics, Harvard T.H. Chan School of Public Health.
  • Denny JC; National Human Genome Research Institute, National Institutes of Health.
  • Hakonarson H; Division of Human Genetics, Children's Hospital of Philadelphia.
  • Mosley JD; Center for Applied Genomics, Children's Hospital of Philadelphia.
  • Smoller JW; Department of Pediatrics, University of Pennsylvania, Perelman School of Medicine.
  • Wei WQ; Department of Medicine, Vanderbilt University Medical Center.
  • Ritchie MD; Department of Biomedical Informatics, Vanderbilt University Medical Center.
  • Moore JH; Psychiatric and Neurodevelopmental Genetics Unit, Center for Genomic Medicine, Massachusetts General Hospital.
  • Chen Y; Center for Precision Psychiatry, Department of Psychiatry, Massachusetts General Hospital.
medRxiv ; 2024 Jan 10.
Article en En | MEDLINE | ID: mdl-38260403
ABSTRACT
Genome-wide association studies (GWAS) have been instrumental in identifying genetic associations for various diseases and traits. However, uncovering genetic underpinnings among traits beyond univariate phenotype associations remains a challenge. Multi-phenotype associations (MPA), or genetic pleiotropy, offer important insights into shared genes and pathways among traits, enhancing our understanding of genetic architectures of complex diseases. GWAS of biobank-linked electronic health record (EHR) data are increasingly being utilized to identify MPA among various traits and diseases. However, methodologies that can efficiently take advantage of distributed EHR to detect MPA are still lacking. Here, we introduce mixWAS, a novel algorithm that efficiently and losslessly integrates multiple EHRs via summary statistics, allowing the detection of MPA among mixed phenotypes while accounting for heterogeneities across EHRs. Simulations demonstrate that mixWAS outperforms the widely used MPA detection method, Phenome-wide association study (PheWAS), across diverse scenarios. Applying mixWAS to data from seven EHRs in the US, we identified 4,534 MPA among blood lipids, BMI, and circulatory diseases. Validation in an independent EHR data from UK confirmed 97.7% of the associations. mixWAS fundamentally improves the detection of MPA and is available as a free, open-source software.
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Tipo de estudio: Prognostic_studies / Risk_factors_studies Idioma: En Revista: MedRxiv Año: 2024 Tipo del documento: Article Pais de publicación: Estados Unidos
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Tipo de estudio: Prognostic_studies / Risk_factors_studies Idioma: En Revista: MedRxiv Año: 2024 Tipo del documento: Article Pais de publicación: Estados Unidos