RepeatedHighDim

RepeatedHighDim: Statistical framework for high-dimensional gene expression and proteomics analysis

RepeatedHighDim implements global testing procedures to evaluate associations between high-dimensional molecular measurements, including gene expression and proteomics data, and experimental factors.

Key Features:

• Global Test Procedures: Adapts global tests originally developed for RNA transcript analysis to high-throughput proteomics datasets.
• Mixed Linear Model: Applies mixed linear models to analyze protein-level data while accounting for incomplete observations.
• Permutation Procedure: Uses non-parametric permutation testing to assess statistical significance and power in complex proteomics data structures.
• Imputation Techniques: Implements imputation strategies for missing values, validated by simulation to reduce bias under defined conditions.
• Gene Ontology (GO) Term Ranking: Ranks GO terms associated with protein sets to evaluate biological processes and pathways.
• Multiple Protein Representation Handling: Aggregates multiple 2-D gel spots corresponding to a single protein into unified protein sets.

Scientific Applications:

• Proteomics Association Analysis: Identifies associations between proteins, including calreticulin, and experimental conditions or biological processes in datasets such as heart muscle tissue studies.

Methodology:

Combines mixed linear modeling, permutation-based significance testing, and missing value imputation to maintain robustness and statistical power in high-dimensional proteomics and gene expression datasets.