Neural Sync Active
Learning Objectives
Registry Synced
Learning Objectives
263 words
1 min read
Reading compass
Now · 1. Gene Regulatory Networks (GRNs)
Learning Objectives
- Understand gene regulation as a network
- Infer regulatory networks from expression data
- Identify network motifs
1. Gene Regulatory Networks (GRNs)
Nodes = genes (including TFs). Edges = regulatory (TF binds promoter -> expression). Directed, can be weighted.
2. Inference Methods
Correlation: Co-expressed -> co-regulated. Simple, no direction. Mutual Information (ARACNE): Information-theoretic, non-linear. Bayesian Networks: Directed, probabilistic, causal structure. Regression (GENIE3): Predict each gene from TF expression.
ChIP-seq: Experimental TF binding sites. Direct evidence.
3. Network Motifs
Recurring patterns with specific functions:
- Feedforward Loop (FFL): A regulates B, both regulate C. Coherent: delay. Incoherent: pulse.
- Feedback Loop: Positive: switch/bistability. Negative: homeostasis.
- Single-Input Module: One TF regulates many targets (coordinated control).
Q1: Correlation vs Bayesian network inference?Correlation: undirected, linear, simple. Bayesian: directed (causal direction), models non-linear relationships. Bayesian needs more data but provides richer model. Q2: What is a feedforward loop?Three-node motif: A->B, A->C, B->C. Coherent FFL: B activates C, creates delay. Incoherent FFL: B represses C, creates pulse. Both perform specific signal processing. Q3: How is ChIP-seq used for GRNs?Crosslink TF to DNA, immunoprecipitate, sequence bound fragments. Identifies genome-wide binding sites. Nearest genes are candidate targets. Q4: What is network inference challenge?High-dim data (20k genes) with few samples (10-100). Correlation not causality. Many models fit same data. Validation needs perturbation experiments. Q5: What is autoregulation?TF regulates own expression. Negative: fast, robust response (homeostasis). Positive: bistability (switch). Join Discord PreviousProtein-Protein Interaction NetworksNextMetabolic Networks