Methods and software tool for complex trait analyses using multi-omics data (2016–2020)
Understanding the genetic architecture of complex traits (including risks to common diseases) is of key importance for medicine and public health in humans. The availability of multi-omics data in large cohorts provides opportunities to quantify systematically the genetic control of DNA methylation, genetic and epigenetic regulations of gene expression, and their effects on complex traits. This project aims to develop novel statistical methods to disentangle the phenotypic variation of a complex trait into components attributable to genomic, methylomic and transcriptomic profiles, to identify the associations of complex traits with DNA methylation or gene expression at particular genomic loci, and to use multi-omics data to predict individuals¿¿¿ phenotypes or risks to diseases in new samples. The project also aims to develop an efficient, versatile, and user-friendly software tool to implement all the commonly used methods, together with those being developed in this project, for complex trait analysis of multi-omics data. The novel methods and powerful software will be of great significance to the identification of novel genes or DNA functional elements involved in complex trait variation, and the use of multi-omics data for risk prediction to prioritize individuals in screening programs.