Back

Genomic and Evolutionary Determinants of Two-hit Frequencies in Tumor Suppressor Genes

Mukherjee, N.; Sabarinathan, R.

2026-02-18 genomics
10.64898/2026.02.17.706175 bioRxiv
Show abstract

While biallelic or "two-hit" inactivation is a central organizing principle for tumor suppressor genes (TSGs), large-scale cancer genomic data reveal substantial heterogeneity in the frequency of such events across genes. This variability reflects diverse selective constraints on allelic disruption, whose biological determinants remain incompletely characterized. Here, we present a comprehensive, allele-specific analysis of TSG two-hit alterations across [~]9,000 tumors from The Cancer Genome Atlas, focusing on loss of heterozygosity (LOH) arising from the co-occurrence of point mutation and deletion. We show that two-hit frequencies vary widely across TSGs and scale with the functional impact and selection strength of point mutations. Integrating mutation position with zygosity reveals distinct patterns consistent with dominant versus recessive modes of action, enabling a zygosity-informed framework for variant interpretation. We further demonstrate that chromosomal context strongly shapes two-hit frequencies, reflecting aneuploidy biases across chromosome arms and selective trade-offs imposed by neighboring loci, including the co-deletion of synergistic TSGs. Extending the LOH analysis beyond diploid tumors, we find that equivalent "all-hit" frequencies are largely preserved in polyploid cancers following whole-genome doubling, consistent with early acquisition of LOH during clonal evolution. Collectively, our results uncover multiple determinants of adherence to the two-hit model, providing new insight into long-standing heterogeneity in TSG behavior and its potential clinical relevance.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.