Table 2.
Software tools available for cancer phylogenetics using CNAs and SNVs.
| Tool | Data | Model | Algorithm | Pros | Cons | Availability |
|---|---|---|---|---|---|---|
| Methods without clonal decomposition | ||||||
| SCARLET | Read count, CNA tree | Dollo | ML | First method applicable to hundreds of cells | Requires CNAs and SNVs from the same set of cells | https://github.com/raphael-group/scarlet |
| Methods with clonal decomposition | ||||||
| PhyloWGS | Read count, clonal allele-specific CN | ISM | Bayesian | First automated method applicable to bulk WGS data of single or multiple samples | Requires subclonal CNAs as input | https://github.com/morrislab/phylowgs |
| Canopy | VAF, fractional allele-specific CN | ISM | Bayesian | Applicable to bulk data from multiple samples | Requires robust data pre-analysis | https://github.com/yuchaojiang/Canopy |
| PhylogicNDT | Purity, read count, allele-specific CN | ISM | Bayesian | Applicable to bulk data from single or multiple samples | Requires robust data pre-analysis | https://github.com/broadinstitute/PhylogicNDT |
| Pairtree | Read count | ISM | Bayesian | Scalable to as many as 30 subclones | Mainly tested on WES data | https://github.com/morrislab/pairtree |
| CONIPHER | Read count, fractional allele-specific CN | ISM | MP | Incorporates error correction | Mainly tested on WES data | https://github.com/McGranahanLab/CONIPHER |
| SPRUCE | VAF, clonal allele-specific CN | IAM | MP | Applicable to bulk data from multiple samples | Requires subclonal CNAs as input | https://github.com/raphael-group/spruce |
| SIFA | Read count | ISM | Bayesian | Integrates SNVs, CNAs, and phylogeny within a single framework for bulk data | Designed for WGS data only | https://github.com/zengliX/SIFA |
| BiTSC2 | Read count | ISM | Bayesian | Integrates SNVs, CNAs, and phylogeny within a single framework for single-cell data | Requires CNAs and SNVs from the same set of cells | https://github.com/ucasdp/BiTSC2 |
| SCsnvcna | SNV genotype, CNA tree | Dollo | Bayesian | Applicable to SNVs and CNAs from independent single-cell data sets | Requires robust data pre-analysis | https://github.com/compbio-mallory/SCsnvcna |
| PACTION | SNV tree, CNA tree | No | MP | Integrates available SNV trees and CNA trees | Requires robust data pre-analysis | https://github.com/elkebir-group/paction |
| Methods for building mutation trees | ||||||
| COMPASS | Read count | Dollo | MAP | Scalable to thousands of cells | Designed for target sequencing data only | https://github.com/cbg-ethz/COMPASS |
The methods developed for single-cell data are shown in italics. BiTSC2: Bayesian inference of tumor clone tree by joint analysis of single-cell single-nucleotide variant and copy number alteration; CN: Copy number; CNA: Copy number alteration; COMPASS: Copy number and mutations phylogeny from amplicon single-cell sequencing; CONIPHER: Correcting noise in phylogenetic evaluation and reconstruction; IAM: Infinite allele model; ISM: Infinite site model; MAP: Maximum a posteriori probability; ML: Maximum likelihood; MP: Maximum parsimony; PACTION: Parsimonious clone tree integration; PhylogicNDT: Phylogic N-dimensional with timing; SCARLET: Single-cell algorithm for reconstructing loss-supported evolution of tumors; SNV: Single-nucleotide variant; SIFA: Subclone identification by feature allocation; SPRUCE: Somatic phylogeny reconstruction using combinatorial enumeration: VAF: Variant allele frequency; WES: Whole exome sequencing; WGS: Whole-genome sequencing.