Hunting for microsatellite instability in long-read data with Owl

Quantification of microsatellite instability on a phased genome from PacBio long reads, by building repeat-length distributions and applying a coefficient of variation.

Microsatellite instability is a key biomarker of mismatch repair deficiency and of response to immunotherapy, yet most existing detection methods are optimised for short-read sequencing and rely on a panel of homopolymer markers. Owl quantifies microsatellite instability from PacBio long-read genomic data, using a genome-wide marker set of more than 140,000 microsatellite repeats of 1-6 bp measured across a phased genome. The tool builds repeat-length distributions at each marker site through a wrap-around alignment algorithm and flags somatic instability using the coefficient of variation. Applied to 131 diverse genomes from the Human Pangenome Reference Consortium, Owl scores ranged from 1.4% to 5.4% of markers exceeding the instability threshold; on cancer cell lines and one diffuse astrocytoma tumour-normal pair, Owl identified six MSI genomes with 10-27% unstable markers, in close concordance with an Illumina DRAGEN MSI assay for the astrocytoma. Motif-level analyses reveal shared enrichment of short homopolymer and A- and AT-rich dinucleotide repeats across MSI cancers.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The interesting shift here is one of scale: moving from a panel of a few homopolymer markers to a genome-wide reference set of more than 140,000 microsatellites, with phasing information on top. Microsatellite instability is mostly an issue in somatic oncology, for immunotherapy eligibility, but the same machinery for measuring repeat length on a phased genome is directly relevant to constitutional genetics, from Lynch syndrome to constitutional mismatch repair deficiency. Concordance with an established assay is nonetheless shown on a single tumour-normal pair: this is a proof of principle, not a diagnostic validation dossier.

Analysis by Dr Thibaut Benquey