Mecan4CNA

Mecan4CNA performs baseline calibration and normalization to improve accuracy of somatic copy number variation (sCNV) analysis in cancer genomes.

Key Features:

• Baseline Calibration: Accurately estimates the baseline representing normal DNA copy number by minimizing errors from sample impurity and measurement bias.
• Level Distance Calculation: Calculates the level distance that quantifies the change corresponding to one additional or missing DNA copy for precise CNV interpretation.
• Noise Reduction: Reduces noise in sCNV profiles to enhance signal clarity and reliability.
• Normalization Across Profiles: Normalizes values across sample profiles to ensure consistent signal scales for comparison and integration.
• High Accuracy: Validated on simulated data with 93% accuracy for baseline estimation and 91% for level distance calculation.
• Improved Sensitivity and Specificity: Applied to The Cancer Genome Atlas (TCGA) data, it increased sensitivity and specificity for detecting focal regions compared to original datasets.
• Integration with Downstream Analyses: Produces normalized outputs that improve the performance of downstream analyses such as GISTIC.

Scientific Applications:

• Cancer CNV Profiling: Enables accurate copy-number profiling in cancer genomics to support interpretation of tumor sCNV landscapes.
• Recurrent sCNV Pattern Detection: Facilitates identification of recurrent sCNV patterns and genomic regions associated with tumorigenesis.
• Focal Region Detection in Cohorts: Improves detection of focal copy-number regions in cohort-scale datasets such as TCGA.
• Input for Downstream Analyses: Provides normalized input to enhance results from downstream tools like GISTIC.

Methodology:

Computational steps explicitly include baseline estimation to represent normal DNA copy number, level-distance calculation to quantify per-copy changes, noise reduction in sCNV profiles, normalization across sample profiles, and validation on simulated data.