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

Area of Science:

• Computational Biology
• Cancer Genomics
• Evolutionary Biology

Background:

• Longitudinal single-cell DNA sequencing (scDNA-seq) provides temporal insights into cancer cell evolution.
• Inferring subclonal trees from scDNA-seq data is crucial for understanding cancer growth, prognosis, and treatment.
• Existing computational tools for subclonal tree inference from longitudinal scDNA-seq data are limited in accuracy and scalability.

Purpose of the Study:

• To introduce scLongTree, a novel computational tool for accurate subclonal tree inference from longitudinal scDNA-seq data.
• To address the limitations of existing tools in terms of accuracy and scalability for analyzing cancer evolution.

Main Methods:

• Development of scLongTree, a computational approach for inferring subclonal evolution.
• Benchmarking scLongTree against state-of-the-art tools (LACE, SCITE, SiCloneFit) using simulated datasets.
• Validation of scLongTree on real-world datasets (SA501 and AML107) with varying cell numbers and mutation counts.

Main Results:

• scLongTree accurately infers subclonal trees from longitudinal scDNA-seq data.
• The tool demonstrates superior performance compared to existing methods on simulated data.
• scLongTree accurately interprets tumor growth on the SA501 dataset and is robust to mutation number variations.
• Scalability is demonstrated on the large AML107 dataset with 4,617 cells.

Conclusions:

• scLongTree is a scalable and accurate computational tool for inferring cancer subclonal evolution from longitudinal scDNA-seq.
• The tool enhances the interpretation of tumor progression and has implications for cancer prognosis and treatment strategies.
• scLongTree is freely available, promoting further research in cancer genomics.