PRESM: personalized reference editor for somatic mutation discovery in cancer genomics

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Accurate somatic mutation detection is vital for cancer research.
• Current methods struggle with read mapping due to germline/somatic mismatches with reference genomes.
• Existing personalized genome tools lack compatibility with downstream mutation detection models.

Purpose of the Study:

• To develop a novel tool, PRESM, for building personalized reference genomes.
• To overcome read mapping obstructions in somatic mutation detection.
• To improve the accuracy of somatic mutation detection in cancer genomes.

Main Methods:

• PRESM integrates germline mutations into a reference genome using a two-step substitution procedure.
• This creates a personalized reference genome with reduced genetic distance to tumor DNA.
• The method maintains positional fidelity through an innovative workaround.

Main Results:

• PRESM reduced false-positive somatic mutation calls by up to 55.5%.
• Facilitated discovery of a novel somatic mutation in PDE1A.
• Achieved accuracy improvements without parameter optimization; PRESM is parameter-free.

Conclusions:

• PRESM effectively enhances somatic mutation detection accuracy by using personalized reference genomes.
• The tool offers a parameter-free solution, ensuring consistent improvements in read mapping and reduced false positives.
• PRESM is compatible with existing statistical models, advancing cancer genome analysis.