Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data

Area of Science:

• Genomics
• Cancer Research
• Bioinformatics

Background:

• Exome/whole-genome sequencing provides somatic mutation data but faces challenges from normal cell contamination, tumor heterogeneity, and data size.
• Determining copy number variations (CNVs) from exome data alone is difficult, often requiring single nucleotide polymorphism (SNP) arrays.
• Existing algorithms can estimate absolute, but not allele-specific, copy number profiles from tumor sequencing data.

Purpose of the Study:

• To develop and validate Sequenza, a software package for analyzing paired tumor-normal DNA sequencing data.
• To estimate tumor cellularity, ploidy, and allele-specific copy number and mutation profiles.
• To assess Sequenza's performance against established methods using The Cancer Genome Atlas (TCGA) data.

Main Methods:

• Developed Sequenza software utilizing paired tumor-normal DNA sequencing data.
• Applied Sequenza and two other algorithms to exome sequencing data from 30 TCGA tumors.
• Compared Sequenza's results with matched SNP array data analyzed by the ASCAT algorithm.

Main Results:

• Sequenza showed strong correlation with SNP/ASCAT for cellularity (r=0.90) and ploidy (r=0.94).
• Sequenza's performance significantly outperformed previously published algorithms.
• The software accurately detected ploidy in simulated samples with as little as 30% tumor content.

Conclusions:

• Exome sequencing data, analyzed with Sequenza, is sufficient for inferring DNA copy number aberrations.
• Sequenza enables accurate estimation of tumor cellularity and copy number profiles.
• This approach reduces reliance on SNP arrays for comprehensive tumor genetic analysis.