MSstatsTMT

MSstatsTMT provides statistical methods for relative protein quantification and differential abundance analysis in tandem mass tag (TMT)-labeled mass spectrometry experiments by explicitly modeling biological and technical variation across multiple TMT mixtures and runs.

Key Features:

• TMT support: Supports relative quantification of proteins in tandem mass tag (TMT)-labeled mass spectrometry experiments.
• Multi-run and multi-mixture handling: Handles experiments spanning multiple TMT mixtures and runs and the associated biological and technical variation.
• Complex experimental designs: Accommodates multiple conditions, biological replicates, technical replicates, and unbalanced designs.
• Statistical modeling: Employs a flexible family of linear mixed-effects models tailored to TMT-based data.
• Technical artifacts and missing values: Models complex patterns of technical artifacts and missing values common in TMT workflows.
• Differential abundance detection: Facilitates detection of differentially abundant proteins with attention to sensitivity and specificity.
• Data-processing integration: Integrates with Proteome Discoverer, MaxQuant, OpenMS, and SpectroMine for downstream statistical analysis.
• Evaluation: Performance has been evaluated using controlled mixtures, simulated datasets, and three diverse biological investigations.

Scientific Applications:

• Relative protein quantification: Quantification of protein abundance across multiplexed TMT-labeled samples.
• Differential abundance testing: Identification of proteins with differential abundance across conditions in TMT experiments.
• Large-scale TMT studies: Analysis of large-scale experiments that require multiple TMT runs or mixtures and complex/unbalanced designs.
• Method benchmarking: Statistical validation and benchmarking using controlled mixtures and simulated datasets.

Methodology:

Implements a flexible family of linear mixed-effects models to model biological and technical variation and to accommodate technical artifacts and missing values in TMT-based proteomics.