search

ZJUFanLab / scRank

A computational method to rank and infer drug-responsive cell population towards in-silico drug perturbation using a target-perturbed gene regulatory network (tpGRN) for single-cell transcriptomic data
GNU General Public License v3.0
24 stars 0 forks source link

What to do when a drug has multiple targets? #1

Closed napaintian closed 2 months ago

napaintian commented 2 months ago

Hello, I would like to ask you if a drug has multiple targets, how to write parameters to find the cell type on which the drug acts?

Lee0498 commented 2 months ago

Hi @napaintian,

Thank you for reaching out and for your question!

When dealing with a drug that has multiple targets, you can specify the drug target genes with the parameter perturbed_target in function rank_celltype(). Here’s an example of what the parameters might look like:

scRank_obj <- rank_celltype(object = scRank_obj, perturbed_target = c('MAP2K1',MAP2K2'))
napaintian commented 2 months ago

There are three target genes for my drug (hubgene). I will report an error after running with the code below. What's going on? hubgene <- c("AKT1","EGFR","CASP3") scRank <- CreateScRank(sce_obj, cell_type = "cell_type_leiden0.5", species = "human",target = hubgene ) scRank <- scRank::Constr_net(scRank) scRank <- scRank::rank_celltype(scRank,perturbed_target = hubgene) Error in scRank::rank_celltype(scRank, perturbed_target = hubgene) : The number of perturbed gene should not be more than two!

Lee0498 commented 1 month ago

Dear napaintian, Thank you for your interest in scRank. In its current version, scRank supports up to two target genes as input. This limitation is due to the inherent complexities of multi-target perturbation, such as modeling synergistic effects, which are challenging to accurately capture. To proceed, we recommend prioritizing your gene list to select the top two targets with the most significant biological signals in your data (like variability).