Analysis of trans eSNPs infers regulatory network architecture

Our work focuses on distal eSNPs to map regulatory interactions between different loci and expose the architecture of the regulatory network defined by such interactions.
We develop novel computational methods and apply them to genetics-genomics data in human.
1) We go beyond pairwise associations to define network motifs. We consider transcripts, each weakly associated to a single main SNP, to expose high confidence regulatory modules structures [1].
2) We extend the framework described in [1] to combine every two basic module structures, i.e., modules composed of two genes, that share the same gene pairs, exposing a bi-fan structure in the human regulatory network [2].
3) We integrate eSNP associations with a PPI network. We show that projecting these interactions onto the PPI network exposes topological properties of eSNPs and their targets, i.e., distance between source and target, degree of the source and degree of the target, and unravels different modes of trans regulation [3].
Overall, our work offers insights concerning the topological structure of human regulatory networks and the role genetics plays in shaping them.

Download: eSNPs_Archirecture


Anat Kreimer: ak2996 - at - Columbia - dot - edu


1. Kreimer, A., et al., Inference of modules associated to eQTLs. Nucleic Acids Res, 2012. 40(13): p. e98.
2. Kreimer, A. and I. Pe'er, Co-regulated transcripts associated to cooperating eSNPs define bi-fan motifs in human gene networks. Accepted for publication in PLOS Genetics, 2014.
3. Kreimer, A. and I. Pe'er, Variants in exons and in transcription factors affect gene expression in trans. Genome Biol, 2013. 14(7): p. R71.