| Title | Predicting transcriptional responses to novel chemical perturbations using deep generative model for drug discovery |
|---|---|
| Description | Understanding transcriptional responses to chemical perturbations is central to drug discovery, but exhaustive experimental high-throughput screening of disease and compound combinations is unfeasible. To overcome this limitation, we present a perturbation-conditioned deep generative model named PRnet for predicting transcriptional responses to novel chemical perturbations that were never experimentally perturbed at bulk and single-cell levels. PRnet screened four compound libraries and generated a large-scale integration atlas of perturbation profiles, including 1) 82 cell lines perturbed by 935 FDA-approved drugs, 2) 88 cell lines perturbed by 4,158 active compounds, 3) 14 CRC cell lines perturbed by 30,456 natural compounds, 4) 6 SCLC cell lines perturbed by 29,670 drug-like compounds and 5) 54 tissues perturbed by 935 FDA-approved drugs. |
| Organism | Homo |
| Data Type | Other Type of Genomic Data |
| Data Accessibility | Open-access |
| BioProject | PRJCA028278 |
| Release Date | 2024-09-15 |
| Submitter | Xiaoning Qi (qixiaoning19s@ict.ac.cn) |
| Organization | Institute of Computing Technology, Chinese Academy of Sciences |
| Submission Date | 2024-07-19 |
HTTP download speed may be slow. It is highly recommended that you download the dataset using a dedicated FTP tool (such as FileZilla Client).
| File ID | File Title | Number/Samples | File Type | File Size | File Suffix | Download |
|---|---|---|---|---|---|---|
| OMIX006910-01 | The Inhibitor and Active Compound Metadata | 1 | Other Type of Genomic Data | 760.34 KB | csv | |
| OMIX006910-02 | The CREEDS Disease Metadata | 1 | Other Type of Genomic Data | 271.17 KB | csv | |
| OMIX006910-03 | The Sciplex Compound Metadata | 1 | Other Type of Genomic Data | 31.73 KB | csv | |
| OMIX006910-04 | The Natura Compound Metadata | 1 | Other Type of Genomic Data | 10.42 MB | csv | |
| OMIX006910-05 | The L1000 Cell Line Metadata | 1 | Other Type of Genomic Data | 10.48 KB | csv | |
| OMIX006910-06 | The Sciplex Cell Line Metadata | 1 | Other Type of Genomic Data | 607 B | csv | |
| OMIX006910-07 | Predicted Signature of Inhibitor and Active Compounds | 1 | Other Type of Genomic Data | 181.25 MB | csv | |
| OMIX006910-08 | Predicted Signature of SCLC Cell Lines | 1 | Other Type of Genomic Data | 104.49 MB | csv | |
| OMIX006910-09 | Predicted Signature of Gtex Tissues | 1 | Other Type of Genomic Data | 11.37 GB | csv | |
| OMIX006910-10 | Predicted Signature of FDA Approved Drugs | 1 | Other Type of Genomic Data | 137.18 MB | csv | |
| OMIX006910-11 | The CREEDS Disease Signature | 1 | Other Type of Genomic Data | 18.52 MB | csv | |
| OMIX006910-12 | The FDA Approved Drug Metadata | 1 | Other Type of Genomic Data | 120.66 KB | csv | |
| OMIX006910-13 | The L1000 Compound Metadata | 1 | Other Type of Genomic Data | 3.73 MB | csv |
| Paper Title | Journal Name | Publish Time | Accession | Citing Type |
|---|---|---|---|---|
| DeepICER: A deep learning framework for predicting compound-induced gene expression profiles | Acta Pharmaceutica Sinica B | 2026-02 | OMIX006910 | Secondary use |
| Predicting transcriptional responses to novel chemical perturbations using deep generative model for drug discovery | Nature Communications | 2024-10 | OMIX006910 OMIX005223 | Deposit |