AncesTree

AncesTree reconstructs clonal evolution and tumor composition from multi-sample DNA sequencing data by factorizing variant allele frequencies (VAFs) to infer lineage relationships among single-nucleotide mutations.

Key Features:

• Clonal Evolution Reconstruction: Formalizes reconstruction of tumor clonal evolution using variant allele frequency (VAF) factorization to identify the lineage and ancestry of individual mutations.
• Combinatorial Characterization: Provides a combinatorial framework to characterize potential solutions for VAF factorization and recognizes that the problem is NP-complete.
• Integer Linear Programming Solution: Solves the VAF factorization problem for error-free data using integer linear programming.
• Probabilistic Error Model: Incorporates a probabilistic model to account for sequencing and measurement errors in real-world data.
• Enhanced Mutation Relationship Identification: Leverages ultra-deep sequencing read counts to improve identification of ancestral relationships between mutations via confident VAF estimates.

Scientific Applications:

• Clonal tree inference: Reconstructs clonal trees and lineage relationships among cancer subpopulations from multi-sample DNA sequencing.
• Tumor heterogeneity and progression analysis: Quantifies tumor heterogeneity and informs temporal progression by estimating clone mixing fractions and evolutionary relationships.
• Mutation ancestry mapping: Determines ancestral relationships between single-nucleotide mutations to map mutational histories within tumors.
• Support for targeted therapy research: Provides clonal composition and inferred evolutionary dynamics that can inform targeted therapy and personalized medicine strategies.

Methodology:

Uses multi-sample DNA sequencing data from the same tumor; frames reconstruction as a VAF factorization problem with combinatorial characterization; applies integer linear programming for error-free scenarios and incorporates a probabilistic error model for noisy sequencing data.