MSAP

MSAP attributes mutational signatures in genomic datasets by applying Non-Negative Least Squares (NNLS) together with a parametric bootstrap to quantify signature contributions and their uncertainties across single and doublet base substitutions, insertions and deletions (indels), and structural variants.

Key Features:

• Non-Negative Least Squares (NNLS) Methodology: Uses NNLS to assign non-negative contributions of reference signatures to individual samples.
• Parametric Bootstrap for Uncertainty Estimation: Applies a parametric bootstrap to quantify statistical uncertainty in signature attributions.
• Comprehensive Mutation Type Support: Supports single and doublet base substitutions, insertions and deletions (indels), and structural variants.
• Reproducibility and Scalability: Implemented as a unified Nextflow pipeline with containerization support for cross-platform reproducibility and scalable execution in high-performance computing environments.
• Validation through Simulations: Validated using simulations incorporating reference COSMIC signatures and randomly generated signatures.

Scientific Applications:

• Mutational signature attribution: Attributes signatures to individual samples to quantify contributions of mutational processes.
• Mutagenic process and DNA damage analysis: Enables analysis of mutagenic processes and DNA damage patterns via signature profiles.
• Cancer genomics: Supports identification of mutational processes relevant to diagnosis, prognosis, and treatment strategies in cancer research.

Methodology:

Computational methods explicitly include Non-Negative Least Squares (NNLS) for attribution, a parametric bootstrap for uncertainty estimation, implementation as a Nextflow pipeline with containerization, and validation via simulations using COSMIC and randomly generated signatures.