Principal Investigator: Dr Jin Liu
National University of Singapore, Duke- NUS Medical School, Center for Quantitative Medicine, 20 College Road Academia, Level 6, Room 57 Singapore 169856
Collaborating lead –
Professor Ching-yu Cheng – Singapore Eye Research Institute
Dr Xiang Zhou – University of Michigan (USA)
Dr Xiang Wan – Hong Kong Baptist University (Hong Kong)
Professor Can Yang – Hong Kong University of Science and Technology (Hong Kong)Tags: 30186, Bayes Approach, big data, GWAS, Integrative Analysis
1a: A rigorous statistical method will be developed to modulate the relationships among multiple GWAS or GWAS with multiple traits, and a scalable algorithm will be developed to handle tons of data from UK biobank.
1b: The proposed method will have significant clinical and public health impact. Investigation of the findings in this study may contribute to the elucidation of biological pathways and novel biomarkers underlying complex traits/diseases which will help individuals’ prevention, diagnosis and treatment of genetic diseases and thus improve population health of our society.
1c: Genetic studies have been developed rapidly over past twenty years, including genome-wide association studies (GWAS). UK biobank data provides us a chance to better understand the mechanism behind diseases with a very large sample size. In this research, we aim at developing statistical models to jointly analyze multiple GWAS or GWAS with multiple traits. The computationally efficient algorithms will be developed to handle multiple GWAS.
1d: In this study, we plan to use the full cohort in UK Biobank.
Project extension – March 2020
A rigorous statistical method will be developed to modulate the relationships among multiple GWAS or GWAS with multiple traits, and a scalable algorithm will be developed to handle tons of data from UK biobank.
Our new included collaborator (a clinician) will be able to help us in a good shape to analyze bulk image data by developed statistical methods. We will develop methods based on the probabilistic model and deep learning to extract features from individuals’ images from MRI in various organs (e.g., brain, liver, pancreas, eye) together with genetic data to predict the related diseases (aging-related diseases, eye, liver and pancreas diseases). The features extracted by deep learning will be data-dependent and one of our collaborators will help us interpret the findings and features extracted.
Last updated Mar 12, 2020