Recent studies have shown that QTL mapping using gene expression as a phenotype may be helpful in the discovery of upstream transcription factors in a gene regulatory network. In this project we intend to use genetic marker data as additional evidence for constructing gene regulatory network. Using this marker data and expression data, we intend to find for each gene, expression Quantitative Trait Loci (eQTLs) in the genome which regulate the expression of the corresponding gene.
Bayesian Networks have been successfully applied to this problem in yeast and others have shown how these probabilistic models can incorporate multiple types of information, such as common sequence motifs and common expression. Our strategy is to develop Bayes Nets that jointly model marker and expression data and apply these models to crosses of inbred mouse strains.
The following diagram shows (a) The representation of conventional interval mapping as a graphical model. For an observed i, all candidate genotype sites j are
considered. (b) Our QTG model of a single regulator-target pair of
genes (regulator is gene j and target is gene i). Subgraphs of
(b) represent (c) cis-, (d) trans-, (e) cis-trans-
cases, and (f) no genotypic effect, corresponding to the conventional
BN. Colored and shaded nodes are observed.
The scatter plot shows an example interaction between MAD2 and TAD2. In this case, simple linear regression on expression levels (blue line using model f) suggests a modest, anti-correlated response and no directionality. By contrast, the green and black points (using model d) represent the two genotypes at the position of MAD2 indicating an allele-specific positive correlation and directional effect from MAD2 to TAD2.
(from Kulp D, Jagalur M (2006). "Causal inference of regulator-target pairs by gene mapping of expression phenotypes." BMC Genomics 7:125 online)
Manjunatha Jagalur
