CellPhy

CellPhy infers phylogenetic trees from single-cell single-nucleotide variants to reconstruct cellular lineage relationships and analyze somatic mutations.

Key Features:

• Maximum Likelihood Framework: Implements maximum likelihood phylogenetic inference for single-cell data.
• Finite-site Markov Genotype Model (16 diploid states): Uses a finite-site Markov genotype model with 16 diploid states to represent genotype evolution.
• Error Modeling: Accounts explicitly for amplification error and allelic dropout common in single-cell genomics.
• Somatic SNV Handling: Models somatic single-nucleotide variants observed in individual cells.
• Integration with RAxML-NG: Implemented as part of the RAxML-NG package for phylogenetic inference.
• Confidence Measures: Reports confidence measurements for inferred phylogenetic trees.
• Scalability and Performance: Demonstrated superior accuracy and computational speed in simulations and can handle datasets of hundreds to thousands of cells.

Scientific Applications:

• Single-cell phylogenetics: Reconstruction of cellular lineage relationships from single-cell SNV data.
• Somatic mutation analysis: Analysis of somatic SNVs at the single-cell level for studies of mutation accumulation and clonal structure.
• Complex cellular populations: Phylogenetic analysis in studies involving heterogeneous cell populations.
• Large-scale single-cell sequencing: Application to large datasets comprising hundreds to thousands of cells from single-cell sequencing experiments.

Methodology:

Performs maximum likelihood inference using a finite-site Markov genotype model with 16 diploid states, incorporates models for amplification error and allelic dropout, and is implemented within the RAxML-NG framework.