Complete gene regulatory network inference workflow from processed single-cell data to regulon discovery and perturbation simulation. Supports RNA-only analysis with pySCENIC (GRNBoost2 + RcisTarget + AUCell) and multiome analysis with SCENIC+ for enhancer-driven GRNs. Includes CellOracle for in silico perturbation simulation.
# pySCENIC (RNA-only path)
pip install pyscenic arboreto ctxcore
# SCENIC+ (multiome path)
pip install scenicplus pycisTopic pycistarget
# CellOracle (perturbation simulation)
pip install celloracle
# Common dependencies
pip install scanpy pandas numpy scipy matplotlib
# Download cisTarget databases (once, ~10 GB each)
# Human: hg38_500bp_up_100bp_down and hg38_10kbp_up_10kbp_down
# Mouse: mm10 equivalents
# From: https://resources.aertslab.org/cistarget/databases/
Input data:
- Processed AnnData (QC'd, normalized, clustered) with .raw containing counts
- TF list for species (from cisTarget resources)
- For SCENIC+: paired scRNA-seq and scATAC-seq (multiome)
Tell your AI agent what you want to do: - "Build gene regulatory networks from my single-cell RNA-seq data" - "Run pySCENIC to find regulons in my scRNA-seq dataset" - "Infer enhancer-driven GRNs from my multiome data with SCENIC+" - "Simulate a TF knockout using CellOracle"
"I have a processed scRNA-seq AnnData with cell type annotations. Run pySCENIC to discover regulons and score their activity."
"Find the key transcription factors driving cell identity in my single-cell data."
"I have paired RNA and ATAC from a 10x multiome experiment. Build enhancer-driven regulatory networks with SCENIC+."
"Use SCENIC+ to identify which enhancers are driving TF regulons in my data."
"Simulate what happens if I knock out MYC in my scRNA-seq data using CellOracle."
"Predict cell state transitions after perturbing key transcription factors."