Related Experiment Video
Updated: Jul 8, 2025
11:02
Detecting Somatic Genetic Alterations in Tumor Specimens by Exon Capture and Massively Parallel Sequencing
Published on: October 18, 2013
19.5K
Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment
Marcos Díaz-Gay1,2,3, Raviteja Vangara1,2,3, Mark Barnes1,2,3
1Department of Cellular and Molecular Medicine, UC San Diego, La Jolla, CA 92093, United States.
Bioinformatics (Oxford, England)
|December 14, 2023
Summary
SigProfilerAssignment accurately assigns mutational signatures to cancer genomes. This computational framework analyzes copy-number signatures and individual somatic mutations, outperforming existing methods.
Area of Science:
- Genomics
- Computational Biology
- Cancer Research
Background:
- Understanding cancer genome evolution relies on analyzing mutational signatures.
- Quantifying the activity of each signature is crucial for this analysis.
Purpose of the Study:
- To introduce SigProfilerAssignment, a novel computational framework for assigning mutational signatures.
- To enable analysis of copy-number signatures and probabilistic assignment to individual somatic mutations.
Main Methods:
- Utilizes a custom implementation of the forward stagewise algorithm for sparse regression.
- Employs nonnegative least squares for numerical optimization.
- Developed as both a desktop and an online tool.
Main Results:
- SigProfilerAssignment successfully assigns all types of mutational signatures to individual samples.
- Demonstrated superior performance compared to four common approaches in analyzing 2700 synthetic cancer genomes.
- The first tool capable of analyzing copy-number signatures and assigning signatures to individual somatic mutations.
Conclusions:
- SigProfilerAssignment provides a robust and accurate method for mutational signature analysis in cancer genomes.
- The framework enhances the understanding of mutagenic processes shaping cancer evolution.
- Offers a valuable tool for cancer genomics research with both desktop and web implementations.
