RNA sequencing (RNA-seq) is a powerful tool for studying gene expression and identifying mutations in cancer cells. However, one challenge researchers face is detecting true cancer-related mutations when a normal, non-cancerous counterpart for comparison is unavailable. Without a matched-normal sample, distinguishing between common genetic variations and tumor-specific mutations becomes difficult.

A recent study led by researchers at the Leibniz-Institute DSMZ presents an improved method for analyzing mutations in breast cancer cell lines using RNA-seq. The researchers applied a variant-calling pipeline to 29 breast cancer cell lines, using filtering strategies to exclude errors, RNA editing sites, and common genetic variants found in public databases such as 1000 Genomes, gnomAD, and dbSNP. Their approach reduced hundreds of thousands of detected mutations down to approximately a thousand high-confidence variants per sample.

Scheme for filter steps in the presented pipeline based on RNA-seq tumour-only breast cancer cell lines

As the majority of germline variants are dispensable and cannot be detected without matched-normal samples, several filter steps were applied to the identified variants. Detailed descriptions are given in the text

To test the accuracy of their method, they compared their filtered variants to known cancer mutations from the Catalogue of Somatic Mutations in Cancer (COSMIC). Their results successfully identified about half of the known COSMIC variants, with missing variants mostly attributed to low read depth in sequencing data. Importantly, all 10 COSMIC cancer gene census variants—a key set of hallmark cancer mutations—were correctly identified, demonstrating the effectiveness of their refined pipeline.

This study highlights the potential of optimized RNA sequencing methods to improve cancer variant detection, even without matched-normal samples. By refining how variants are filtered, researchers can gain clearer insights into the genetic landscape of tumor cells, ultimately aiding in cancer diagnostics and drug development.