COMPILE: a GWAS computational pipeline for gene discovery in complex genomes

Area of Science:

• Plant genetics
• Bioinformatics
• Computational biology

Background:

• Genome-Wide Association Studies (GWAS) identify genes contributing to quantitative traits.
• Complex genetic architectures pose computational challenges for gene discovery.
• A streamlined computational pipeline, COMPILE, was developed to accelerate maize gene identification.

Purpose of the Study:

• To accelerate the identification and annotation of candidate maize genes associated with quantitative traits.
• To match maize genes to their closest rice and Arabidopsis homologs.
• To identify candidate loci contributing to European Corn Borer resistance in maize.

Main Methods:

• Developed COMPILE, a computational pipeline for GWAS.
• Utilized a Mixed Linear Model incorporating population structure control.
• Linked significant Quantitative Trait Loci (QTL) to candidate genes and RNA regulatory elements.
• Validated COMPILE using published data for α-tocopherol biosynthesis and flowering time.

Main Results:

• COMPILE identified known and novel candidate genes and non-coding RNAs for α-tocopherol biosynthesis and flowering time.
• Applied to the maize Goodman Association Panel, COMPILE identified candidate loci for European Corn Borer resistance.
• Identified candidate genes including transcriptional factors, signaling molecules, and metabolic enzymes.
• Found a gene of unknown function, homologous to a pathogen-stress-associated gene in rice.

Conclusions:

• COMPILE's speed enables analysis of population size and compression effects.
• Population size and diversity, not marker density, are key constraints for trait analysis.
• COMPILE is customizable and adaptable for other species with comprehensive genomic data.