PathFinder

PathFinder reconstructs migration routes of cancer cell clones by integrating clone phylogenies and mutational data to infer seeding events between primary and metastatic tumors.

Key Features:

• Bayesian Inference: Employs a Bayesian approach to reconstruct migration routes using clone phylogenies within patients, assessing mutational differences among clones and their presence or absence across primary and metastatic tumors.
• Phylogenetic Analysis: Leverages clone phylogeny and the number and distribution of genetic variants to trace migration paths from primary tumors to new metastases and among existing metastases.
• Error Correction: Identifies challenges in tracing backward migrations (from metastases to primary tumors) and indicates that increasing the number of sampled clones per tumor and analyzed genetic variants mitigates these errors.
• Comprehensive Inference: Reconstructs migration routes and provides posterior probabilities for inferred paths to assess the likelihood that new metastases were seeded from primary tumors or existing metastatic sites.

Scientific Applications:

• Metastasis Research: Provides detailed reconstructions of cancer cell migration histories to study how and when metastases form.
• Clinical Diagnostics: Infers migration routes that can inform clinical strategies for early intervention by identifying likely seeding relationships between tumors.
• Genomic Studies: Enables exploration of intra-patient genetic diversity and the evolutionary dynamics of cancer cell populations for personalized medicine research.

Methodology:

Implemented as a Python script (Windows 64-bit) requiring specific Python packages and no compilation; integrates clone phylogenies with mutational data to model migration history and computes posterior probabilities for inferred migration routes.