search

aertslab / pySCENIC

pySCENIC is a lightning-fast python implementation of the SCENIC pipeline (Single-Cell rEgulatory Network Inference and Clustering) which enables biologists to infer transcription factors, gene regulatory networks and cell types from single-cell RNA-seq data.
http://scenic.aertslab.org
GNU General Public License v3.0
420 stars 179 forks source link

question on positive vs. negative regulons #75

Closed dschrein closed 5 years ago

dschrein commented 5 years ago

I have a situation where the same TF shows up as both a positive and negative regulon in a given cluster of my cells (e.g. Prdm1, Runx2 in the red cluster). Does it just mean that some of its targets are upregulated and some are downregulated? How would I investigate further given the following output files?

adjacencies.p auc_mtx.csv df.p modules.p motifs.csv regulons.p

heatmaps are based on cluster-averaged auc matrix values

image

thanks in advance!

bramvds commented 5 years ago

Dear,

Indeed theoretically this means that this TF has two (disjoint) sets of targetomes: a set of target genes which the factor controls as a transcriptional activator (+ regulon) and a disjoint set of target genes for which the factor acts as a transcriptional repressor (- regulon). However, this relationship between a factor and a target genes is inferred from correlation patterns only. We briefly assessed the relevance of "negative regulons" via the well-known transcriptional repressor REST in a relevant dataset (Zeisel et al.) and decided to focus only on the "positive" regulons. Because of this, by default, pySCENIC only returns "positive" regulons.

Hope this helps, Bram