Options for GRN with single cell

[lpantano]

scMTNI: integrates scRNAseq and scATACseq [Elizabeth]

• https://www.nature.com/articles/s41467-023-38637-9
• Probabilistic model using Bayesian framework, can model covariance between genes
• Cell lineage tree assumes related cell types will have similar GRNs
• Assumes that gene expressions follow a Gaussian model
• Prefers but does not require ATACseq data
• Outputs GRNs for each cell type on a cell lineage tree
• https://github.com/Roy-lab/scMTNI
  • 1 — install: clone from Github, run in Linux environment, requires requires GCC version of gcc-6.3.1 and GNU extension with std=gnu++14 setting
  • 2 — prepare input files: integrate scRNA+scATAC data using LIGER, generate prior network from scATAC, prepare cell lineage tree
  • 3 — run
  • 4 — evaluate
  • 5 — network dynamics analysis: edge-based k-means clustering or Latent Dirichlet Allocation (LDA)-based topic models

TRIPOD: R package, it would give you TF-Gene pairs, it can be ran in specific genes, there are examples of visualization the prediction. https://github.com/yuchaojiang/TRIPOD/tree/main [Lorena]