PolySTest

PolySTest performs robust statistical testing to detect differentially regulated proteins and peptides in mass spectrometry-based quantitative proteomics while explicitly addressing missing values to improve confidence and sensitivity.

Key Features:

• Robust differential testing: Performs statistical tests for high-confidence detection of differentially regulated proteins and peptides in quantitative proteomics datasets.
• Missing-value-aware Miss test: Implements the Miss test that simultaneously evaluates missingness and feature abundance to rescue features affected by incomplete observations.
• Integrated statistical suite: Integrates a suite of statistical tests tailored to handle complexities inherent in mass spectrometry data.
• Platform-independent confidence scores: Produces confidence scores that are robust across instrument platforms, experimental protocols, and upstream software tools.

Scientific Applications:

• Benchmarking with ground truth: Validated on artificial and experimental proteomics datasets with known ground truth, demonstrating increased sensitivity and confidence versus conventional methods.
• Large-scale muscle differentiation proteomics: Applied to mass spectrometry-based proteomics of differentiating muscle cells and rescued 10–20% additional proteins within molecular networks relevant to muscle differentiation.

Methodology:

Combines a suite of statistical tests with a novel Miss test that simultaneously evaluates missingness and feature abundance and is integrated with traditional statistical tests to rescue incomplete features and boost confidence and sensitivity.