Identifying cancer cells from calling single-nucleotide variants in scRNA-seq data.

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Single-cell RNA sequencing (scRNA-seq) is crucial for studying cancer cell heterogeneity.
• Distinguishing tumor cells from healthy cells in scRNA-seq data is challenging due to mixed populations.
• Somatic single-nucleotide variants (SNVs) can identify cancer cells but are difficult to call accurately from scRNA-seq data.

Purpose of the Study:

• To develop an interpretable tool, CCLONE, for identifying cancer cell populations using SNVs from scRNA-seq data.
• To address the challenges of noisy and sparse SNV data in scRNA-seq.
• To jointly identify cancer clones and their associated variants.

Main Methods:

• Developed CCLONE (Cancer Cell Labelling On Noisy Expression), an algorithm designed for noisy and sparse SNV data.
• CCLONE integrates variant calling with clonal population identification.
• The tool is adapted to handle uncertainties inherent in scRNA-seq data.

Main Results:

• CCLONE successfully identified genetic clones and somatic events in multiple patient datasets, including acute myeloid leukemia and lung adenocarcinoma.
• The tool demonstrated its capability to capture complex clonal structures.
• Results highlight CCLONE's effectiveness in analyzing scRNA-seq data for cancer research.

Conclusions:

• CCLONE provides a robust method for identifying cancer cells and their genetic makeup within heterogeneous tumor microenvironments.
• The tool offers valuable insights into cancer cell origins and disease progression.
• CCLONE enhances the utility of scRNA-seq data for cancer genomics research.