Premium
Proximity Model for Expression Quantitative Trait Loci (eQTL) Detection
Author(s) -
Gelfond Jonathan A. L.,
Ibrahim Joseph G.,
Zou Fei
Publication year - 2007
Publication title -
biometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.298
H-Index - 130
eISSN - 1541-0420
pISSN - 0006-341X
DOI - 10.1111/j.1541-0420.2007.00778.x
Subject(s) - expression quantitative trait loci , quantitative trait locus , biology , computational biology , genetics , genome , locus (genetics) , gene , single nucleotide polymorphism , genotype
Summary Expression quantitative trait loci (eQTL) are loci or markers on the genomes that are associated with gene expression. It is well known to biologists that some ( cis ) genetic influences on expression occur over short distances on the genome while some ( trans ) influences can operate remotely. We use a log‐linear model to place structure on the prior probability for genetic control of a transcript by a marker locus so that the loci that are closest to a transcript are given a higher prior probability of controlling that transcript to reflect the important role that genomic proximity can play in the regulation of expression. This proximity model is an extension of the mixture over marker (MOM) model for the simultaneous detection of cis and trans eQTL of Kendziorski (Kendziorski et al., 2006, Biometrics 62 (1), 19–27). The genomic locations of the transcripts are used to improve the accuracy of the posterior distribution for the location of the eQTL. We compare the MOM method to our extension with both simulated data and data sets of recombinant inbred mouse lines. We also discuss an extension of the MOM method to model multiple eQTLs, and find that many transcripts are likely associated with more than one eQTL.