|
Abstract:
|
Complex traits like brain disorders and cancers are highly polygenic, often involving hundreds of genes and multiple biological processes. GWAS have identified hundreds of loci robustly associated with the disease risk for a variety of diseases. Translating the GWAS findings to mechanisms and ultimately therapeutics, however, has proven challenging, largely due to the lack of mapping of GWAS loci to underlying genes. For such complex diseases, disease-causing genes often exhibit aberrations from multiple levels, reflected in multi-omics data (e.g. transcriptome, methylation, Hi-C data for chromatin interaction). In this talk I will describe Bayesian model selection frameworks that integrate gene-gene networks and multi-omics data to implicate underlying disease-causing genes and biological processes, with the rationale that each of the multi-omics data provides partial and complementary information, integration of which increases the precision to pinpoint the disease genes. I will show that the identified disease-causing genes hold great potential for drug development and repositioning, for brain disorders and cancers.
|
