Detecting Somatic Mutations Without Matched Normal Samples Using Long Reads.

Area of Science:

• Genomics
• Cancer Biology
• Bioinformatics

Background:

• Somatic mutation detection is vital for personalized cancer treatment.
• Standard mutation calling requires matched normal samples, which are not always available.
• Existing methods for normal-matched-free calling have limitations in accuracy or data requirements.

Approach:

• Utilizes single-molecule long read sequencing for somatic mutation identification without matched normal samples.
• Develops a theoretical framework demonstrating the utility of haplotype phasing in distinguishing somatic mutations.
• Employs simulations to evaluate performance across varying tumour purity and sequencing depths.

Key Points:

• Haplotype phasing provides critical information for accurate somatic mutation calling.
• The smrest tool, a prototype caller, is developed based on this approach.
• Demonstrated efficacy on highly mutated cancer cell lines.

Conclusions:

• Single-molecule long read sequencing offers a viable alternative for mutation calling when normal samples are absent.
• This method has the potential to accurately measure genome-wide mutation biomarkers, including tumour mutation burden and mutation signatures.