Personalized genome assembly for accurate cancer somatic mutation discovery using tumor-normal paired reference...

Area of Science:

• Genomics
• Cancer Research
• Bioinformatics

Background:

• Personalized, haplotype-specific genome assemblies are advocated for cancer mutation detection but lack exploration in tumor-normal paired samples.
• Existing reference genomes like GRCh38 are mosaic and may limit the detection of the full spectrum of somatic events.

Purpose of the Study:

• To demonstrate the first use of a de novo assembled personalized genome as a reference for cancer mutation detection.
• To quantify the impact of reference genomes on the accuracy of somatic mutation detection in paired tumor-normal samples.

Main Methods:

• Generated de novo assemblies of the first tumor-normal paired genomes (nuclear and mitochondrial) from an individual with triple-negative breast cancer.
• Utilized chromosomal-scale, haplotype-phased, and annotated personalized genomes for analysis.
• Compared mutation detection accuracy using personalized versus standard reference genomes.

Main Results:

• Personalized genomes provide individual-specific haplotypes, improving read alignments for both short-read and long-read sequencing data.
• Enhanced accuracy in detecting somatic single nucleotide variants (SNVs) and structural variants (SVs).
• Identified novel somatic mutations exclusively when using the personalized genome assembly as the reference.

Conclusions:

• Personalized genomes with individual-specific haplotypes are crucial for accurate somatic mutation detection in paired tumor-normal samples.
• The developed resource and methodology will advance precision oncology for breast and other cancers.