Characterizing and modeling the genomewide molecular basis of gene-environment interactions

PROJECT SUMMARY/ABSTRACT Organismal phenotypes in a given environment frequently differ from what might be expected based on genotypic or environmental data alone. These genotype-specific deviations, or gene-environment interactions (GxE), can constitute a large portion of phenotypic variation and are important for determining an individual’s

wellbeing in its given environment. An individual adapted to a particular environment can respond appropriately to typical local stresses and nutrients, but may be maladapted in new or changing environments. GxE also makes it exceedingly difficult to predict organismal response to the environment: the magnitude and direction

of GxE effects depend on the loci, alleles, traits, and environments involved. This complicates extrapolation of genomic prediction models into new populations or environments. Although it is well established that GxE is a major contributor to phenotypic variation, much less is known about the molecular mechanisms determining

individuals’ differential response to environments. This is particularly true in complex, real world environments that are impossible to reproduce in laboratory experiments. Genomewide, allelic variation for gene expression cumulatively influences GxE of organism-level phenotypes, but the complex networks and patterns of gene

regulation driving GxE are not well understood. Over the coming five years, this project will generate new datasets and analyze existing datasets to begin understanding and modeling the genomewide patterns of gene expression that cumulatively determine GxE in real world environments. In Aim 1, tissue samples

from multi-environmental experiments will be used to evaluate the landscape of gene expression among genetically diverse individuals grown in a variety of environments. Specifically, we will investigate how changes to cis-regulatory sequences (e.g. transcription factor binding motifs) contribute to GxE for gene expression.

Simultaneously, we will identify genes that show GxE for expression levels and model how they contribute to GxE for organism-level phenotypes. In Aim 2, we will use existing datasets independent yet complementary to those generated in Aim 1 to test whether GxE in organism-level phenotypes can be predicted directly from

sequence variation. Together the multi-scale projects in this study range from the sequence level to the entire organism. By studying GxE at multiple scales and with a variety of different data types, this study will strengthen our understanding of how allelic sequence variation changes gene regulatory networks and drives

local adaptation. These findings are important for understanding how organisms adapt to new environments and for better predicting organismal response to the environment.