yamss

yamss performs preprocessing and analysis of high-throughput metabolomics data obtained by chromatography-mass spectrometry to reduce peak quantification variability and increase statistical power for differential analysis.

Key Features:

• Analysis and visualization: Supports analysis and visualization of high-throughput metabolomics data acquired by chromatography-mass spectrometry.
• Advanced preprocessing methodology: Implements the bakedpi preprocessing method using intensity-weighted bivariate kernel density estimation that pools all samples during peak detection.
• Reduction of variability: Pooled peak detection reduces unnecessary variability in peak quantifications compared with per-sample preprocessing.
• Enhanced analytical power: Improved preprocessing increases the statistical power of downstream differential analysis, enabling detection of subtle metabolic differences.
• Compatibility with multiple data modes: Handles both centroid and profile mode mass spectrometry metabolomics data.

Scientific Applications:

• Biomedical research: Supports biomarker discovery and metabolic pathway elucidation in mass spectrometry-based metabolomics studies.
• Differential analysis: Enables more reliable differential analysis to compare metabolic profiles across conditions or treatments by reducing variability in peak quantification.

Methodology:

bakedpi applies an intensity-weighted bivariate kernel density estimation and pools all samples during initial peak detection instead of processing each sample independently, thereby minimizing quantification variability.