Knet

Knet identifies and analyzes pairs of covariant genes across gene expression datasets to assess changes in gene network topology associated with disease progression, including metastatic transitions in solid tumors.

Key Features:

• Gene Pair Identification: Identifies pairs of covariant genes that exhibit consistent expression patterns across datasets, including in control states with replicate variability.
• Covariance Analysis: Detects gene pairs that co-vary even when individual gene expression is unstable across biological replicates.
• Network Topology Analysis: Examines how identified gene pairs interact and form networks to reveal changes in connectivity between normal and pathological conditions.
• Disease Progression Modeling (DGPMs): Implements Decoherence Gene Pair Models (DGPMs) to capture transitions of network topology from healthy to diseased states, including shifts from richly connected to sparser configurations during metastatic transition.
• Comparative Network Assessment: Compares the size and structure of gene networks under different biological conditions to quantify topology differences.
• Validation and Testing: Applies models in vitro and validates findings across a range of solid tumor studies to assess generalizability and correlations with survivability.
• Control and Test Condition Handling: Operates on gene expression data collected from both control and test conditions with biological replicates.

Scientific Applications:

• Cancer progression analysis: Analyzes network topology changes in solid tumors and metastatic transitions to inform studies of disease mechanism and prognosis.
• Biomarker and therapeutic-target discovery: Reveals evolving gene interactions during metastatic transition to support identification of candidate biomarkers or therapeutic targets.
• Comparative condition studies: Enables quantitative comparison of gene network size and structure between control and disease states.

Methodology:

Gather gene expression data from control and test conditions with biological replicates; perform covariance analysis to identify covariant gene pairs; develop Decoherence Gene Pair Models (DGPMs) to model coherence and network topology changes; apply in vitro validation and test models against solid tumor studies.