Regulatory Network Inference (pySCENIC)

This page documents the pySCENIC-based regulon analysis pipeline. See Methods for the manuscript prose.

Pipeline

Step 1 — Regulon Inference (GRNBoost2)

  • Input: raw-count scRNA-seq expression matrix
  • Runs: 40 independent GRNBoost2 runs
  • Pruning: motif-based using motifs-v9-nr.hgnc (m0.001)
  • CisTarget database: hg38 RefSeq r80 ±10 kb TSS (hg38__refseq-r80__10kb_up_and_down_tss.mc9nr)
  • Aggregation: regulons retained if recovered in ≥ 25% of runs

Step 2 — Activity Scoring (AUCell)

  • Input: ComBat-seq–corrected pseudobulk profiles
  • Purpose: reduce sparsity and improve robustness

Step 3 — Differential Testing

  • Per cell type and ancestry: regulon activity compared between T2D and healthy
  • Test: Wilcoxon rank-sum test
  • Correction: FDR (Benjamini–Hochberg)

Software

  • pySCENIC v0.12.0