Although genome-wide association study (GWAS) has been a great success in the past 15 years in identifying genomic regions associated with complex diseases and traits, major challenges remain both for identifying functional genes and variants in statistically significant regions, as well as in interpreting GWAS results. These challenges are due to the low signal noise ratios in GWAS analysis, complex dependence among genetic markers, and the lack of access to individual level data for many studies.

In this lecture, Professor Hongyu ZHAO will first introduce a statistical model that is commonly used to characterise the genetic contributions to complex traits and its robustness to model misspecifications. He will then describe its extensions to identify relevant tissues/cell types for a specific trait using genome annotations and estimate genetic correlations (both global and local) between different traits. Professor Zhao will also discuss statistical inference using either individual genotype and phenotype data, a typical set up for traditional statistical analysis, or summary statistics, which are more easily accessible from GWAS. 

For more information, please visit the announcement page here.