Pangenome Graph Node-Phenotype Association shows GWAS-like quality results with only few individuals

GraNPA identifies known phenotype-linked loci in pangenome variation graphs built from only 13 Oryza sativa genomes and 24 Bos taurus genomes, matching expected regions.

• Why it matters: Association studies on PVGs often need many individuals and can suffer from reference bias in variant calling; this method targets qualitative phenotypes with minimal sampling.
• What they did: The approach scores graph nodes for phenotype–sequence association within a PVG, then detects statistically significant shifts in node score distributions to localize trait regions.
• The result: In simulated contexts and two public datasets, it recovered the expected Sub1A region in a 13-individual rice PVG and the white-headed insertion locus in a 24-individual cattle PVG.

AI summary of the abstract · may contain errors · not medical advice · read the paper.