Approved research

Novel methods for detecting single-marker associations, gene-gene and gene-environment interactions, with a focus on the X chromosome

Lay summary

We have developed a suite of powerful statistical methods for detecting genetic associations, gene-gene and gene-environment interactions, and for X-linked genetic associations. We aimed to apply these new methods on existing data sets from UKBiobank. We are especially interested in autoimmune diseases, mental disorders, cardiovascular diseases, and their related biomarkers. These diseases and traits are sexually dimorphic, suggesting the potential role of genetic variations on the X chromosome. Our novel methods have the potential to unravel new genetic associations, gene-gene and gene-environment interactions, and X-linked genetic associations for clinically relevant traits. Although current genome-wide association studies have identified many associated genetic markers, most of them exhibit small effect sizes and they collectively explain only a small proportion of the heritability. The missing heritability may lie in gene-gene interaction and on the X chromosome. Our study will increase our understanding of the genetic basis of relevant traits/diseases, improving our future prevention, diagnosis and treatment of illness. This proposed project will perform a series of statistical analysis on existing data from UK Biobank. In this regard, statistical methods previously developed by our group, specifically for gene-gene and gene-environment interactions, and for X-centered association analysis, will be applied on the genotype data and relevant phenotypic data from UK Biobank. Independent of UK Biobank data, upon the discovery of novel statistically significant associations, we will interpret the biological significance and functional relevance of the associations with other public databases or with additional molecular experiments in human cells or model organisms. We apply to include all participants with relevant traits (autoimmune diseases, mental disorders, cardiovascular diseases, and their related biomarkers). A big sample size will increase the power for detecting associations.