plrs

plrs models gene-level relationships between DNA copy number and mRNA expression using Piecewise Linear Regression Splines to identify gene-specific associations.

Key Features:

• Flexible Modeling Framework: Uses Piecewise Linear Regression Splines instead of fixed parametric forms to model associations between DNA copy number and mRNA expression.
• Interpretable Models: Provides a range of interpretable models, including commonly used variants, enabling incorporation of prior biological knowledge.
• Gene-Specific Relationship Identification: Identifies the specific type of relationship between gene copy number and mRNA expression on a per-gene basis.
• Statistical Rigor: Models associations and tests their strength, reporting confidence intervals for estimates.
• Parallel Computation Support: Includes code for parallel computation to process large datasets such as those from The Cancer Genome Atlas (TCGA).

Scientific Applications:

• Cancer genomics: Characterizes the interplay between DNA copy number alterations and gene expression to reveal oncogenic mechanisms.
• TCGA glioblastoma analysis: Applied to glioblastoma data from TCGA to uncover gene-level copy number–expression associations that may inform therapeutic hypotheses.

Methodology:

Uses Piecewise Linear Regression Splines to model DNA copy number versus mRNA expression, capturing non-linear relationships and breakpoints, performing statistical tests of associations with confidence intervals, and supporting parallel computation for large datasets.