SRMstats

SRMstats implements statistical analysis for protein quantification from Selected Reaction Monitoring (SRM) targeted mass spectrometry experiments using linear mixed-effects models to assess protein-level significance across isotopic labels, peptides, charge states, transitions, samples, and experimental conditions.

Key Features:

• Statistical Modeling Framework: Employs a family of linear mixed-effects models that integrate quantitative measurements across isotopic labels, peptides, charge states, transitions, samples, and experimental conditions.
• Protein Significance Analysis: Detects proteins with differential abundance between conditions while controlling the false discovery rate.
• Experimental Design Support: Supports isotope label-based and label-free SRM workflows and accommodates group comparison and time-course experimental designs.
• Validation: Accuracy and reproducibility verified using controlled datasets from the NCI-CPTAC reproducibility investigation and an internal spike-in study.
• Implementation: Provided as an R package and designed for integration into computational pipelines.

Scientific Applications:

• Biomarker Discovery: Quantifies targeted proteins to identify candidates with differential abundance across conditions.
• Clinical Diagnostics: Measures changes in protein levels relevant to disease states for diagnostic studies.
• Proteomic Profiling: Profiles protein abundance across biological conditions, time courses, and experimental groups.
• SRM Experimental Design Optimization: Informs design and analysis choices for isotope label-based and label-free SRM experiments.

Methodology:

Applies linear mixed-effects models integrating measurements across isotopic labels, peptides, charge states, transitions, samples, and experimental conditions, with statistical testing that controls the false discovery rate; validated on NCI-CPTAC reproducibility and internal spike-in datasets and implemented in R.