scLongTree: an accurate computational tool to infer the longitudinal tree for scDNAseq data

Abstract

AbstractA subclonal tree that depicts the evolution of cancer cells is of interest in understanding how cancer grows, prognosis and treatment of cancer.Longitudinal single-cell DNA sequencing data (scDNA-seq) is the single-cell DNA sequencing data sequenced at different time points. It provides more knowledge of the order of the mutations than the scDNA-seq taken at only one time point, and thus facilitates the inference of the subclonal tree. There is only one existing tool LACE that can infer a subclonal tree based on the longitudinal scDNA-seq, and it is limited in accuracy and scale.We presented scLongTree, a computational tool that can accurately infer the longitudinal subclonal tree based on the longitudinal scDNA-seq. ScLongTree can be scalable to hundreds of mutations, and outper-formed state-of-the-art methods SCITE, SiCloneFit and LACE on a comprehensive simulated dataset. The test on a real dataset SA501 showed that scLongTree can more accurately interpret the progres-sive growth of the tumor than LACE. ScLongTree is freely available onhttps://github.com/compbio-mallory/sclongitudinal infer.