|
Activity Number:
|
317
- Integration Approaches and Methods for Deciphering Genotype-Phenotype Mapping Toward Precision Medicine
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #322211
|
|
|
Title:
|
Polygenic Risk Modeling Techniques to Incorporate LD, Functional Annotations of SNPs And/Or Multiple Phenotypic Information Using Genome-Wide Association Study Summary-Level Data
|
|
Author(s):
|
Ting-Huei Chen* and Nilanjan Chatterjee and Jianxin Shi
|
|
Companies:
|
Université Laval and Johns Hopkins University and National Cancer Institute
|
|
Keywords:
|
Polygenic risk scores ;
GWAS ;
complex diseases ;
summary-level data
|
|
Abstract:
|
Polygenic risk scores (PRS) has been a popular tool for genetic risk prediction of complex diseases. The modelling techniques working on summary-level data would avoid the difficulties in the policy issues of data sharing. The typical summary-level data is the statistics on the association parameter in marginal analysis on each single nucleotide polymorphism (SNP) from GWAS. The standard PRS is based on a set of independent SNPs with p-values less than a predefined significance level. However, this approach may discard important information and reduce the prediction power. Thus one important challenge is to re-estimate the parameters of correlated SNPs. In addition, recent studies have suggested that the incorporation of external information may ameliorate the performances of PRS. We propose new PRS modeling techniques working on summary-level data based on the penalized estimation to incorporate the correlation structure of SNPs, functional annotations and/or multiple phenotypic information. Simulation results demonstrate good performance of the methods. The real data analysis illustrates the practical utility of the techniques.
|
Authors who are presenting talks have a * after their name.
