Leveraging human pangenome for improved somatic variant detection.

Graph-based human pangenomes and personalized pangenome-inferred donor-specific assemblies improve somatic variant detection accuracy versus GRCh38 by better alignment and inference.

• Why it matters: Somatic calling is hindered by low variant allele fractions, germline variation contamination, and reference bias from linear references that miss sample-specific sequence.
• What they did: The authors benchmarked GRCh38, graph-based pangenomes, and pangenome-inferred DSAs using a HapMap mixture dataset and the COLO829 melanoma cell line, assessing alignment and variant calling performance.
• The result: Pangenome-guided alignment boosts mapping and somatic variant calling accuracy, while personalized pangenomes partially reconstruct donor-specific content, reduce germline contamination, and enable detection in loci absent or poorly represented in GRCh38.

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