TargetClone: A multi-sample approach for reconstructing subclonal evolution of tumors

Area of Science:

• Oncology
• Computational Biology
• Genetics

Background:

• Tumors exhibit significant subclonal heterogeneity, complicating the study of cancer progression and treatment response.
• Accurate reconstruction of tumor evolutionary relationships is essential but challenging due to sample heterogeneity and difficulties in obtaining unbiased read depth measurements.
• Microdissection coupled with targeted sequencing offers a way to reduce heterogeneity, but requires specialized computational tools.

Purpose of the Study:

• To introduce TargetClone, a novel computational method for reconstructing tumor subclonal evolution trees.
• To enable accurate inference of copy numbers, alleles, and tumor fraction from targeted sequencing data.
• To provide a tool specifically designed for microdissected samples to overcome limitations in existing methods.

Main Methods:

• Developed TargetClone, a Python-based computational method utilizing single-nucleotide polymorphism allele frequency and somatic single-nucleotide variant data.
• Designed TargetClone to reconstruct subclonal evolution trees and infer genomic features like copy number and tumor fraction.
• Validated TargetClone on simulated data and real-world datasets from testicular germ cell and ovarian cancers.

Main Results:

• TargetClone demonstrated low error rates on simulated data, indicating high accuracy.
• The method successfully reconstructed expected evolutionary trees in complex cancer datasets.
• TargetClone provides essential insights into tumor heterogeneity and subclonal architecture.

Conclusions:

• TargetClone is an effective tool for deconvolving tumor heterogeneity from targeted sequencing data.
• The method facilitates a deeper understanding of tumor evolution, progression, and response to therapy.
• TargetClone is publicly available, promoting further research in cancer genomics and personalized medicine.