Integrating Coexpression Networks with GWAS to Prioritize Causal Genes in Maize

Abstract

Genome-wide association studies (GWAS) have identified loci linked to hundreds of traits in many different species. Yet, because linkage equilibrium implicates a broad region surrounding each identified locus, the causal genes often remain unknown. This problem is especially pronounced in nonhuman, nonmodel species, where functional annotations are sparse and there is frequently little information available for prioritizing candidate genes. We developed a computational approach, Camoco, that integrates loci identified by GWAS with functional information derived from gene coexpression networks. Using Camoco, we prioritized candidate genes from a large-scale GWAS examining the accumulation of 17 different elements in maize (Zea mays) seeds. Strikingly, we observed a strong dependence in the performance of our approach based on the type of coexpression network used: expression variation across genetically diverse individuals in a relevant tissue context (in our case, roots that are the primary elemental uptake and delivery system) outperformed other alternative networks. Two candidate genes identified by our approach were validated using mutants. Our study demonstrates that coexpression networks provide a powerful basis for prioritizing candidate causal genes from GWAS loci but suggests that the success of such strategies can highly depend on the gene expression data context. Both the software and the lessons on integrating GWAS data with coexpression networks generalize to species beyond maize.

Multi-allelic APRR2 Gene is Associated with Fruit Pigment Accumulation in Melon and Watermelon

“…A key challenge remains in prioritizing GWAS signals and improving weak signals obtained from low frequency causative variants (Lee and Lee, 2018). The availability of whole-genome assemblies and corresponding protein-coding gene annotations, alongside additional layers of information, such as expression profiles from RNA-Seq experiments, now facilitate the integration of multiple data layers to improve GWAS results (Shim et al , 2017; Lee and Lee, 2018; Schaefer et al , 2018). Our example of the CmAPRR2 gene suggests that adding functional annotation prediction to GWAS SNPs and treating predicted genes as integral functional units could potentially be used as an informative layer that can boost the signal of causative weak associations.…”

The evolutionary forces shaping cis and trans regulation of gene expression within a population of outcrossing plants

“…Therefore, the coexpression modules we observed were shaped by genetic perturbations, not tissue or developmental differences. While coexpression measured across multiple timepoints ('developmental networks') has been linked to functional relationships (Eisen et al 1998;Stuart et al 2003) , coexpression modules generated from genetically distinct individuals have different properties than those generated from different tissue types (Mähler et al 2017;Schaefer et al 2018) . In some cases, this difference is helpful: analyses combining GWAS and coexpression networks have the most power when using coexpression networks made from genetically distinct samples (Schaefer et al 2018) .…”

Genome-wide Association Studies of Agronomic Traits Consisting of Field- and Molecular-based Phenotypes

“…These information may lead to increased efficiency of crop breeding, and accelerate the development of new cultivars for sustainable food production. gene co-expression network analysis is a powerful approach toward understanding the molecular processes controlling the phenotype (Kobayashi et al, 2016;Li et al, 2018;Schaefer et al, 2018), in which large expression data would be acquired from accessions under GWAS than public database to explain the GWAS phenotype. Recently, public databases for plant phenotype have been established (Arabidopsis: AraPheno [Seren et al, 2017], rice: RiceVarMap [Zhao et al, 2015], Maize: Panzea [Zhao et al, 2006]).…”