A k-mer-based genome-wide association study approach empowering gene mining in polyploids

Area of Science:

• Genomics
• Plant Breeding
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) in polyploids face challenges due to genotyping ambiguity and allele dosage complexity.
• Existing methods struggle to accurately analyze the genetic variations in complex polyploid genomes.

Purpose of the Study:

• To introduce KMERIA, a novel k-mer-based framework designed to overcome genotyping and association mapping limitations in polyploid genomes.
• To enable efficient and robust genetic analysis for complex polyploid species.

Main Methods:

• Developed KMERIA, a k-mer-based computational framework for polyploid genotyping and association mapping.
• Integrated KMERIA with a graph pangenome approach to account for structural variations.
• Applied the framework to 290 wild sugarcane (Saccharum spontaneum) accessions.

Main Results:

• KMERIA demonstrated superior accuracy and statistical power compared to existing methods in benchmarking tests.
• Identified novel genes associated with sucrose biosynthesis (SsMGT) and tillering (e.g., SsERF14, SsNGA5, SsNAC, SsARF8, SsLOG, SsSCR) in sugarcane.
• Elucidated the genetic architecture of yield-related traits in wild sugarcane.

Conclusions:

• KMERIA effectively addresses critical methodological gaps in polyploid genomics.
• The integration of KMERIA with graph pangenomes provides a powerful approach for genotype-phenotype relationship studies in complex crops.
• Findings offer actionable targets for improving sugarcane breeding programs.