search

sankaranlab / SCAVENGE

SCAVENGE is a method to optimize the inference of functional and genetic associations to specific cells at single-cell resolution.
GNU General Public License v3.0
80 stars 34 forks source link

Potential for scRNA-seq? #1

Closed cnk113 closed 2 years ago

cnk113 commented 2 years ago

Hello,

I was wondering if the model's assumption holds for scRNA-seq data?

Best, Chang

fl-yu commented 2 years ago

Hi Chang,

Thank you for your interest in our tool. It is totally feasible for scRNA-seq data.

I think two points could be modified for this purpose:

  1. Select the seed cells. We use g-chromVAR to calculate an enrichment Z score by combining posterior probability and scATAC-seq peak signal for each individual cell, and the top enriched cells are selected as seed cells. You probably need to replace this by using a RNA/scRNA-seq orientated method for this task.
  2. Graph construction. We used LSI (or batch effect corrected LSI) matrix from scATAC-seq for graph construction. You may need other approaches like PCA from scRNA-seq for this task.

Then it would be feasible for SCAVENGE analysis from scRNA-seq. We are still actively developing this tool, a future direction is scalable to scRNA-seq, stay tuned!