Title

STIP: An SNP-trait inference platform

Document Type

Article

Publication Date

2017

Keywords

Bayes methods, Bioinformatics, Genomics, Diseases, Frequency control, belief networks, DNA, genetics, medical computing, Bayesian network, STIP, SNP-trait inference platform, SNP genotypes, SNP-categorical trait associations, SNP-quantitative trait associations, Top-k trait prediction, genome-wide association studies, genotype inference, single nucleotide polymorphisms, GWAS statistics, GWAS catalog expression, GWAS catalog

Abstract

Genome-wide association studies (GWASs) have received increasing attention to understand how a genetic variation affects different human traits. Recent works show that the Bayesian network is powerful in modeling the conditional dependency between single-nucleotide polymorphisms (SNPs) and traits using only the GWAS statistics. In this paper, we present STIP, a web-based SNP-trait inference platform capable of a variety of inference tasks, such as trait inference given SNP genotypes and genotype inference given traits. The core of STIP is two Bayesian networks which model the SNP-categorical trait associations and SNP-quantitative trait associations, respectively. Both Bayesian networks are derived from the public GWAS catalog. The inference tasks are based on the dependency relationship captured in the Bayesian networks. The current version of STIP provides three services which are SNP-trait inference, Top-k trait prediction and GWAS catalog exploration.

Comments

Principal Investigator: Xintao Wu

Acknowledgements: This work is supported in part by U.S. National Science Foundation (DGE-1523115 and IIS-1502273). We also would like to thank Dr. Xinghua Shi from UNC Charlotte for helpful discussions and feedback.

Share

COinS