rHVDM

rHVDM predicts transcription factor activity and potential targets from gene expression time-course data to elucidate regulatory mechanisms underlying microarray-derived expression changes.

Key Features:

• R implementation: Implemented in R for analysis of gene expression time-course data.
• Dynamic mechanistic model: Uses a dynamic, mechanistic model of gene expression to represent transcription factor regulation over time.
• Two-step inference: Derives a TF activity profile from a limited set of known target genes and uses that profile to predict additional putative targets.
• Measurement-error modeling: Incorporates measurement error into the predictive framework to provide an objective assessment of prediction robustness and quality.
• Computational efficiency: Employs efficient algorithms and vectorization to accelerate optimization and differential equation integration.
• Microarray focus: Designed to address challenges posed by highly parallel genomic platforms such as microarrays and their differential expression lists.

Scientific Applications:

• Transcription factor activity inference: Infers temporal activity profiles of transcription factors from time-course expression data.
• Target prediction: Predicts putative transcription factor targets based on inferred activity profiles.
• Regulatory dynamics analysis: Analyzes dynamic regulatory interactions, exemplified by studies of the tumor suppressor p53.
• Interpretation of differential expression: Helps interpret extensive lists of differentially expressed genes from microarray experiments by linking them to regulatory mechanisms.

Methodology:

Employs a dynamic mechanistic model of gene expression; uses a two-step procedure to derive TF activity from a limited set of known targets and then predict additional targets; incorporates measurement-error modeling; and performs optimization and differential equation integration using vectorized efficient algorithms.