dynDeepDRIM: a dynamic deep learning model to infer direct regulatory interactions using time-course single-cell gene expression data
Problem
Reconstructing cell-type-specific gene regulatory networks from time-course single-cell RNA-seq data, which existing tools could not do while handling multiple time points and scRNA-seq characteristics.
Why it matters
Gene regulatory networks control transcription factor activity in cells, and reconstructing them is key to understanding dynamic gene expression changes.
What the AI did
dynDeepDRIM, a deep learning model, represents joint expression of a gene pair as an image and uses the target TF-gene pair image plus neighbor images to reconstruct GRNs and remove transitive TF-gene interactions.
Results
not stated in source
Impact
Provides a method for GRN reconstruction from time-course scRNA-seq data, per the abstract.
Record
| Venue | Briefings in Bioinformatics |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Yu Gong Xu, Jiaxing Chen, Aiping Lyu, William K. Cheung, Lu Zhang |
| Citations | 23 (OpenAlex, at import time) |
| References | 59 |
| Model or method | dynDeepDRIM (deep learning model) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1093/bib/bbac424 |
| Open access | green |
| Keywords | gene regulatory networks, single-cell RNA-seq, deep learning, time-course, transcription factors |
Importance rating
real but narrower advance
New method for cell-type-specific GRN reconstruction from time-course scRNA-seq; useful for understanding gene expression dynamics but abstract gives no scale or clinical impact.