Redefined indel taxonomy reveals insights into mutational signatures

Despite their deleterious effects, small insertions and deletions (indels) have received far less attention than substitutions1,2. Recent computational advances have surmounted previous technical challenges, enabling the study of mutational processes driving indel formation. Here we generated isogenic CRISPR-edited human cellular models of post-replicative repair dysfunction (PRRd), including individual and combined gene-edits of mismatch repair (MMR) and replicative polymerases (POLE and POLD1). Unique, diverse mutational footprints of MMR deficiency and polymerase dysfunction were revealed. However, the prevailing indel classification framework1 falls short in discriminating these indel signatures from background mutagenesis and among each other, as it condenses important biological signals into predominantly two indel subclasses, limiting signature analyses. To address this, we propose a novel classification system that considers the 5’ and 3’ nucleobases flanking an indel, as with substitution signatures, resulting in 89 indel subclasses. Our new indel classification enhances the disambiguation of experimental signatures, and through analysis of 4,775 cancer whole genomes from the 100,000 Genomes Project3, we uncover 37 indel signatures (InDs); 27 are new. Further, we develop a classifier – PRRDetect, which outperforms approved biomarkers, such as Tumor Mutational Burden (TMB), for immunotherapy. This re-defined indel taxonomy advances our understanding of mutagenesis and heralds potential clinical applications.

Deposited here are the de novo mutation lists from the experimental samples included in the study. Whole-genome sequencing short reads were aligned to GRCh38/hg38 using BWA-MEM. Post-processing filters were applied to improve the specificity of mutation-calling. Specifically, for single nucleotide variant calls by CaVEMan, we used CLPM == 0 and ASMD >= 140. To reduce false positive calls by Pindel, we used QUAL >= 250 and REP < 10. Rearrangements were not assessed as they were too few to be informative. De novo substitutions and indels in subclones were obtained by subtracting from respective parental clone whenever available, or by removing mutations shared among subclones.