HATCHet2: clone- and haplotype-specific copy number inference from bulk tumor sequencing data

Area of Science:

• Genomics
• Cancer Biology
• Bioinformatics

Background:

• Bulk DNA sequencing of multiple tumor samples is increasingly common.
• Existing methods often analyze tumor samples independently, limiting comprehensive CNA analysis.
• Accurate identification of copy-number aberrations (CNAs) is crucial for understanding tumor evolution.

Purpose of the Study:

• To introduce HATCHet2, an advanced algorithm for simultaneous haplotype- and clone-specific CNA identification from multiple bulk tumor samples.
• To enhance the detection of focal CNAs and introduce a novel statistic for identifying mirrored-subclonal CNAs.
• To improve the accuracy of CNA inference in cancer genomics.

Main Methods:

• Development of HATCHet2, an extension of the HATCHet algorithm.
• Introduction of the minor haplotype B-allele frequency (mhBAF) statistic.
• Validation using simulations and a single-cell sequencing dataset.

Main Results:

• HATCHet2 demonstrates improved accuracy in identifying focal CNAs compared to previous methods.
• The novel mhBAF statistic enables the detection of previously elusive mirrored-subclonal CNAs.
• Analysis of 10 prostate cancer patients revealed new mirrored-subclonal CNAs impacting cancer genes.

Conclusions:

• HATCHet2 provides a more accurate and comprehensive approach to analyzing CNAs across multiple tumor samples.
• The method facilitates the discovery of complex subclonal CNA events, including mirrored CNAs.
• HATCHet2 has significant implications for cancer genomics research and understanding tumor heterogeneity.