| Title: | Identifying gene networks underlying complex traits in the UK Biobank |
| Lead Institution: | Brown University |
| Principal investigator: | Dr Sohini Ramachandran |
Application 22419
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Patients with complex diseases can have different mutations within a single gene, or set of interacting genes, which predispose them to the same disease. To address this issue, we are developing a new statistical method that produces ?gene scores? summarizing how strongly genes and pathways are associated with a phenotype. We propose to apply our method to multiple traits assayed in the UK Biobank, including Schizophrenia, Bipolar disorder, Type 2 Diabetes, Crohn?s Disease, Ulcerative Colitis, Rheumatoid Arthritis, Standing Height and Body-Mass Index. Our research will yield new insight into the genetic mechanisms underlying these traits. We are developing a new method to identify candidate genes and pathways that may represent potential drug targets and treatments for complex phenotypes. Our method will be freely available and easy-to-use so that it may be applied to health-related studies by other researchers. In the long term, the results of our proposed research project can be used to better understand how complex diseases are inherited, which will improve clinical decision-making for patients that are genetically predisposed to multiple phenotypes. Our new method PEGASUS calculates gene-based association scores that take into account genomic correlations resulting from genomic linkage and population history. We will first calculate genotype-level association scores for the following traits in the UK Biobank: Schizophrenia, Bipolar disorder, Type 2 Diabetes, Crohn?s Disease, Ulcerative Colitis, Rheumatoid Arthritis, Standing Height and Body-Mass Index. We will then perform gene-level tests of association using PEGASUS. We can validate our results using previously published results for the same traits. We will also perform pathway and gene network analyses using PEGASUS to characterize interacting genes that underlie each phenotype. We would like to include the full cohort of participants. Approximately 16000 individuals will serve as cases in our association analyses: Schizophrenia (637 cases), Bipolar Disorder (1462 cases), Type 2 Diabetes (3682 cases), Crohn?s Disease (1544 cases), Ulcerative Colitis (2757 cases), and Rheumatoid Arthritis (5893 cases). The remaining cohort individuals will be controls. We will also perform association analyses on Standing Height and Body-Mass Index (~500000 individuals). The genotype data for the full cohort are essential for us to accurately calculate genotypic correlations that are then used to determine the null distribution for our gene-level association test.3 Publications
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 7816 | Enabling interpretable machine learning for biological data with reliability scores | K. D. Ahlquist (+2) | 2023 | PLOS Computational Biology |
| 7163 | Enrichment analyses identify shared associations for 25 quantitative traits in over 600,000 individuals from seven diverse ancestries | Samuel Pattillo Smith (+14) | 2022 | American Journal of Human Genetics |
| 10321 | Uncertainty quantification in variable selection for genetic fine-mapping using bayesian neural networks | Wei Cheng (+2) | 2022 | iScience |
Enabling scientific discoveries that improve human health