Age-dependent topic modeling of comorbidities in UK Biobank identifies disease subtypes with differential genetic risk
Year: 2023
Type: article
Abstract: Longitudinal data from electronic health records (EHR) has immense potential to improve clinical diagnoses and personalised medicine, motivating efforts to identify disease subtypes from age-dependent patient comorbidity information. We introduce an age-dependent topic modelling (ATM) method that provides a low-rank representation of longitudinal records of hundreds of distinct diseases in large EHR data sets. The model learns, and assigns to each individual, topic weights for several disease topics, each of which reflects a set of diseases that tend to co-occur as a function of age. Simulations show that ATM attains high accuracy in distinguishing distinct age-dependent comorbidity profiles. We applied ATM to 282,957 UK Biobank samples, analysing 1,726,144 disease diagnoses spanning 348 diseases with ≥1,000 incidences. We inferred 10 disease topics optimising model fit. We identified 52 diseases with heterogeneous comorbidity profiles (≥500 incidences assigned to each of ≥2 topics), including breast cancer, type 2 diabetes (T2D), hypertension, and hypercholesterolemia; for most of these diseases, topic assignments were highly age-dependent, suggesting differences in disease aetiology for early-onset vs. late-onset disease. We defined subtypes of the 52 heterogeneous diseases based on the topic assignments, and compared genetic risk across subtypes using polygenic risk scores (PRS). We identified 18 disease subtypes whose PRS differed significantly from other subtypes of the same disease, including a subtype of T2D characterised by cardiovascular comorbidities and a subtype of asthma characterised by dermatological comorbidities. We further identified specific SNPs underlying these differences. For example, the T2D-associated SNP rs1063192 in the CDKN2B locus has a higher odds ratio in the top quartile of cardiovascular topic weight (1.19±0.02) than in the bottom quartile (1.08±0.02) (P=4×10-5 for difference). In conclusion, ATM identifies disease subtypes with differential genome-wide and locus-specific genetic risk profiles.
Source: Nature Genetics
Institutions: Centre for Human Genetics, Harvard University, University of Cambridge, British Heart Foundation, University of Oxford, Broad Institute, Chinese Academy of Medical Sciences & Peking Union Medical College, Peking Union Medical College Hospital, Baker Heart and Diabetes Institute, The Alan Turing Institute, Health Data Research UK
Language: English
Subfield: Genetics
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $11,690
Funders: U.S. Department of Health and Human Services, Harvard University, Li Ka Shing Foundation, Harvard T.H. Chan School of Public Health, Wellcome Trust, Alan Turing Institute, Broad Institute, UK Research and Innovation, National Institute for Health and Care Research, British Heart Foundation, Department of Health and Social Care, Research Councils UK, University of Oxford, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Peking Union Medical College, Baker Heart and Diabetes Institute, National Institutes of Health, Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Biomedical Research Council, NIHR Oxford Biomedical Research Centre, NIHR Cambridge Biomedical Research Centre, National Institute of Mental Health, National Human Genome Research Institute, Health Data Research UK
Awards: G1199782147, G139696651, Illinois Precision Medicine Consortium , G1579515842, G1671907826, Healthy Americas: All of Us Research Program, G1969433312, G1974824942, G2187930703, G2266788633, Large-scale integrative genomic studies of high-dimensional traits in cardiovascular diseases (renewal), Empowerment-focused Outreach, Education and Enrollment of Tennessee Older Adults in the All of Us Research Program through Community Efforts by FiftyForward, G2465463101, G2537968379, Research Matters: Creating Possibilities to Achieve Health and Wellness for All of Us Research Program, G3177290858, Research Excellence (round 3), G3402893575, G370598735, Precision Medicine Initiative Cohort Program Biobank, Communications Design and Development for NIH Precision Medicine Initiative Research Projects, New York City Consortium for Precision Medicine, Leveraging functional data to predict disease risk in multi-ethnic populations, G4629282264, G4753630297, G4763310767, New Approaches to Bayesian Data Science: Tackling Challenges from the Health Sciences, Technology to Empower Changes in Health (TECH) Network Participant Technologies Center, G5421189793, University of Arizona-Banner Health All of Us Research Program , The New England Precision Medicine Consortium of the All of Us Research Program, G5850737185, Discovering genetic drivers of longitudinal electronic health records, G6001630035, All of Us Wisconsin , Data and Research Support Center , All of Us Pennsylvania, G639689439, Trans-America Consortium of the Health Care Systems Research Network for the All of Us Research Program, G6631254222, Artificial intelligence to create equitable multi-ethnic polygenic risk scores that improve clinical care, G6641991021, Southern All of Us Network, The BHF Chair of Epidemiology and Medicine, SouthEast Enrollment Center (SEEC) , G7523544532, Large-scale integrative studies of risk factors in coronary heart disease: from discovery to application (renewal). (Joint funding with MRC), Adaptive Platform for Personalized Engagement, Research Excellence, G8355735039, California Precision Medicine Research Program Consortium, Communication Implementation and Coordination for NIH Precision Medicine Initiative Research Projects, G9000046247, G9000286330, Statistical methods to localize disease heritability and identify biological mechanisms, Functional and population genetic architectures of complex disease, PRIDEnet for the All of Us Research Program
OpenAlex ID: W4387451594