Incorporating ENCODE Information into SNP-Based Phenotype Prediction

• Activity Number: 353
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Yue-Ming Chen* and Peng Wei
• Companies: The University of Texas at Houston and The University of Texas School of Public Health
• Keywords: GWAS ; ENCODE ; functional annotations ; phenotype prediction

Abstract

Recent studies show that most genome-wide association studies (GWAS) identified single nucleotide polymorphisms (SNPs) fall outside of the protein-coding regions (Hindorff et al. 2009). These trait-associated SNPs may present regulatory function and tend to be highly correlated, i.e., in high linkage disequilibrium (LD), with the functional SNPs (Maurano et al. 2012, Schaub et al. 2012). Tools for annotating functional variation in human genome integrate enriched regulatory information from multiple resources such as the Encyclopedia of DNA Elements (The ENCODE Project Consortium 2012). Using the ENCODE information, we performed functional annotation of SNPs, and we constructed a weighted genetic prediction framework for complex traits. Two frameworks were considered as prototypes for incorporating the ENCODE information: the weighted penalized regression (Zou 2006) and the weighted false discovery rate (Genovese et al. 2006). Using simulations and real data analysis, we compared the weighted procedures to the unweighted procedures and to an extended best linear unbiased prediction method (Speed and Balding 2014).