Gene Regulatory Network Inference
A gene regulatory network is a directed graph in which an edge from one gene to another means the first, typically a transcription factor, controls the expression of the second. Inferring such networks from expression measurements is a long-standing problem because expression data are correlational: co-expressed genes may share a regulator, respond to the same stimulus, or simply share technical properties, so any inference method must be judged against an evidence source that distinguishes regulation from co-occurrence.
Questions this Concept answers
- Why can an expression-only gene regulatory network confuse a true regulatory edge with an indirect one, such as two genes sharing a common upstream driver?
Inferring Gene Regulatory Networks from Expression Data in Computational Biology
Computational biology infers gene regulatory relationships by combining probabilistic reasoning (Bayesian networks over hidden causal variables and observed data), distance/similarity metrics (Euclid…