Inferring and summarizing tumor phylogenies from bulk DNA data.

Area of Science:

• Computational Biology
• Genomics
• Evolutionary Biology

Background:

• Cancer phylogenies are crucial for understanding tumor evolution.
• Phylogenetic estimation from bulk sequencing data often yields multiple plausible trees, hindering downstream analyses.
• Accurate cancer phylogenies are essential for personalized medicine and therapeutic strategies.

Purpose of the Study:

• To introduce Sapling, a novel computational method for inferring backbone trees from cancer sequencing data.
• To address the challenge of phylogenetic uncertainty in tumor evolution studies.
• To develop a method that summarizes the space of plausible cancer phylogenies.

Main Methods:

• Sapling solves two variants of the Backbone Tree Inference from Reads problem.
• The method identifies a concise set of backbone trees representing a subset of mutations.
• The computational complexity of the underlying problems was proven to be NP-hard.

Main Results:

• Sapling effectively infers high-quality backbone trees that summarize plausible cancer phylogenies.
• The method demonstrates robust performance on both simulated and real tumor sequencing data.
• Sapling successfully infers full-size phylogenetic trees with higher likelihoods compared to existing state-of-the-art approaches.

Conclusions:

• Sapling provides a robust solution for inferring backbone trees in cancer phylogenetics.
• The method enhances the analysis of tumor evolution by managing phylogenetic uncertainty.
• Sapling offers improved accuracy in reconstructing cancer phylogenies, advancing the field of computational oncology.