Conbase: a software for unsupervised discovery of clonal somatic mutations in single cells through read phasing

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Accurate variant calling and genotyping are critical for single-cell genomics.
• Current methods face limitations in identifying somatic mutations in single-cell DNA sequencing data.

Purpose of the Study:

• To introduce Conbase, a novel computational tool for somatic mutation identification in single-cell DNA sequencing.
• To enhance the confidence and accuracy of somatic variant calls and genotype predictions.

Main Methods:

• Conbase utilizes phased read data from multiple samples within a dataset.
• The method is designed to improve the reliability of somatic variant detection.

Main Results:

• Conbase demonstrated superior performance compared to three other methods.
• It achieved the best results in terms of false discovery rate and specificity.
• The tool showed enhanced robustness on simulated data, in vitro expanded fibroblasts, and primary lymphocyte populations.

Conclusions:

• Conbase offers a significant advancement for somatic mutation identification in single-cell genomics.
• The method provides increased confidence and accuracy for variant calling and genotyping.
• Conbase is a robust tool applicable to various single-cell DNA sequencing datasets.