Mathematical Sciences Research Seminar Series
POLARIS: Polygenic LD-Adjusted Risk Score Set-Based Method
As part of the School of Science and Technology Mathematical Sciences Research seminar series, Emily Baker, MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, presents: POLARIS: Polygenic LD-Adjusted Risk Score Set-Based Method .
- From: Wednesday 3 May 2017, 3 pm
- To: Wednesday 3 May 2017, 4 pm
- Location: New Hall Block NHB169, Clifton campus, Nottingham Trent University, Clifton Campus, Clifton Lane, Nottingham, NG11 8NS
- Download this event to your calendar
As part of the School of Science and Technology Mathematical Sciences Research seminar series, Emily Baker, MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, presents: POLARIS: Polygenic LD-Adjusted Risk Score Set-Based Method.
Genome-wide association studies (GWAS) investigate whether genetic variations (SNPs) in individuals are associated with disease. Single SNPs have a small effect on disease and therefore power to detect disease associations can be gained by using the combined effect of SNPs within a set, such as a gene or the whole genome. Polygenic risk scores are a weighted score used to capture the combined effect of a set of SNPs, and can increase the power of set-based analyses by leveraging large public GWAS datasets.
We propose the application of polygenic risk scores as a set-based method with an additional component of adjustment for linkage disequilibrium (correlation between SNPs); this informs the analysis with previously reported effect sizes of a SNP’s association to disease, and accounts for linkage disequilibrium between SNPs.
We call this method POLARIS: POlygenic Linkage disequilibrium-Adjusted RIsk Score. POLARIS identifies the linkage disequilibrium structure of SNPs using spectral decomposition of the SNP correlation matrix and adjusts the effect sizes used in the risk score. POLARIS scores are calculated per subject per set and the overall association of the set is determined using logistic regression on the adjusted polygenic risk score and additional population covariates.
We used simulations to compare the power of POLARIS to Polygenic Risk Score and other set based approaches which also account for linkage disequilibrium between markers. We then applied these approaches to real data to check the consistency with simulations. We observed that POLARIS has greater power than other set-based methods in the majority of datasets considered.
This seminar will be hosted by Professor Nadia Chuzhanova.
For any enquiries please contact Dr Jonathan Crofts.