Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dmis-lab/badger
https://github.com/dmis-lab/badger
Last synced: about 1 month ago
JSON representation
- Host: GitHub
- URL: https://github.com/dmis-lab/badger
- Owner: dmis-lab
- Created: 2024-02-19T01:35:06.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-06-18T04:59:35.000Z (6 months ago)
- Last Synced: 2024-06-18T05:59:07.874Z (6 months ago)
- Language: Python
- Size: 44.9 KB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# BADGER: Biologically-Aware Interpretable Differential Gene Expression Ranking Model
## Abstract
Understanding which genes are significantly influenced by a drug can reveal insights into the mechanism of action, a vital aspect of drug repurposing. A drug affecting specific pathways or gene expressions in one disease could potentially be effective in another with similar genetic patterns. Ranking genes according to the extent of their expression change within cells, pre- and post-drug treatment, allows us to identify the genes most substantially affected by a particular drug. However, previous studies' limited scope of cells and explainability constraints hinder a comprehensive understanding of drug-cell response. We introduce BADGER, a Biologically-Aware interpretable Differential Gene Expression Ranking model. This model is designed to predict gene expression changes resulting from interactions between cancer cell lines and chemical compounds. It employs a similarity-based method for representing novel and diverse cancer cell lines. Additionally, the three attention blocks in the model mimic the cascading effects of chemical compounds, ensuring a thorough consideration of their complex interactions with cancer cell lines. Moreover, the integration of prior knowledge about drugs' target into the model enhances its explainability. Comparative evaluations demonstrate that BADGER outperforms baseline models in capturing the intricate interaction between cancer cell lines and chemical compounds. Its application in drug repurposing is further validated by analyzing the attention maps and predicted rankings of differentially expressed genes for both approved and repurposing candidate drugs, highlighting its potential in identifying novel therapeutic uses for existing drugs. (*submitted to Bioinformatics, under review*)## Dataset
You can download the preprocessed dataset from this [google drive link](https://drive.google.com/drive/folders/19-qR-TDAKOAc00_nSU4IIu63IWFnZ4YJ?usp=sharing).
## Contributors
Name
Affiliation
Hajung Kim†
Data Mining and Information Systems Lab,
Korea University, Seoul, South Korea
[email protected]
Mogan Gim†
Data Mining and Information Systems Lab,
Korea University, Seoul, South Korea
[email protected]
Seungheun Baek
Data Mining and Information Systems Lab,
Korea University, Seoul, South Korea
[email protected]
Soyon Park
Data Mining and Information Systems Lab,
Korea University, Seoul, South Korea
[email protected]
Sunkyu Kim*
AIGEN Sciences, Seoul, South Korea
[email protected]
Jaewoo Kang*
Data Mining and Information Systems Lab,
Korea University, Seoul, South Korea
[email protected]
- †: *Equal Contributors.*
- *: *Corresponding Author*