search

aertslab / SCENICprotocol

A scalable SCENIC workflow for single-cell gene regulatory network analysis
GNU General Public License v3.0
139 stars 62 forks source link

Will regulon outputs be biased by a large proportion of homogeneous cells in my dataset? #81

Open katsturgess opened 9 months ago

katsturgess commented 9 months ago

Hi, I would like to run PyScenic on a 10X dataset of 195,000 cells we have generated from mouse bone marrow. The dataset is overall quite diverse in terms of number of celltypes, but around 25% of cells in the bone marrow are a single celltype, Neutrophils. Will this bias the analysis to reduce sensitivity to regulons which are not upregulated or downregulated in the Neutrophil lineage? And if so could I downsample the Neutrophil clusters to reduce their influence on the overall regulon inference? Thanks for your advice.