Table 1.

List of methods for tumour phylogeny inference from scDNAseq data, with their main features

Method	SNVs	CNAs	Doublets	SNV Recurrence	SNV loss	Homozygous mutations	Est. max # cells	Est. max # loci
∞SCITE7, 8	Yes	No	Yes	Yesa	Yesa	No	10,000	100
SCIΦN9, 10	Yes	No	No	Yes	Yes	No	100	1000
OncoNEM11	Yes	No	No	No	No	No	100	100
SiCloneFit12	Yes	No	Yes	Yes	Yes	No	100	100
SPhyr13	Yes	No	No	No	Yes	No	100	100
SCICoNE14	No	Yes	No	–	–	–	100	–
CHISEL15	Yesb	Yes	No	–	–	–	1000	–
SCARLET16	Yes	Noc	No	No	Yesd	No	100	100
BiTSC217	Yes	Yese	No	No	Yes	Yes	100	100
COMPASS	Yes	Yes	Yes	No	Yesf	Yesf	10,000	100

The maximum number of cells and loci are estimates for reasonable runtimes and performance.

^aHowever model selection is not automated.

^bCan assign SNVs to clones after the CNA-tree is inferred by aggregating all cells assigned to each clone.

^cRequires CNA tree as input, which must be obtained with another method.

^dIf supported by copy-number loss; which could miss CNLOH.

^eAssumes that all loci have the same coverage (in the absence of CNAs), which is not the case for targeted sequencing.

^fWith copy number loss or CNLOH.