Confounding factors in assessing the enriched expression of somatic mutant allele in bulk tumor samples

Abstract

Allele-specific expression (ASE) of somatic mutations can be caused by cis-activation of the mutant allele or silencing of the wildtype allele and has been investigated by examining the enrichment of mutant allele in RNA relative to DNA. Here we show that this mutation-based approach can be confounded by gene expression differences in tumor and normal cells which co-exist in most bulk tumor samples. We modeled mutant allele expression by incorporating tumor/normal expression difference, mutant allele dosage, tumor purity, and nonsense-mediated decay (NMD) efficiency, projecting that the enrichment can occur without ASE. This confounding effect is exacerbated with low tumor purity and is dependent on mutant allele dosage for NMD-sensitive mutations. The model was validated empirically using somatic insertions/deletions (indels) from 9,101 the Cancer Genome Atlas (TCGA) samples, showing three-fold higher enrichment in driver genes vs. non-drivers. To explore if the enriched mutant allele expression can be leveraged to complement DNA-based somatic mutation detection, we performed de novo somatic indel calling using TCGA RNA-Seq, increasing the TCGA driver indel repertoire by [~]14%, mostly in samples with lower tumor purity. Our study not only revealed confounding factors in ASE analysis but also demonstrated the utility of leveraging RNA-Seq for driver mutation analysis.