Anti-Dengue: A Machine Learning-Assisted Prediction of Small Molecule Antivirals against Dengue Virus and Implications in Drug Repurposing
Problem
Dengue virus has no approved antivirals, so computational methods are needed to predict small-molecule inhibitors and repurposed drug candidates.
Why it matters
Dengue outbreaks persist in global tropical regions lacking approved antiviral treatments.
What the AI did
Built the 'Anti-Dengue' QSAR/ML pipeline computing PaDEL descriptors and fingerprints on DrugRepV chemicals, training regression models (SVM, ANN, kNN, RF) with 10-fold cross-validation to predict dengue inhibitors.
Results
Best model PCC of 0.71 on training/testing dataset and 0.81 on independent validation dataset.
Impact
Provides a predictive model for dengue inhibitor candidates; no experimentally confirmed antiviral is stated.
Record
| Venue | Viruses |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Sakshi Gautam, Anamika Thakur, Akanksha Rajput, Manoj Kumar |
| Citations | 30 (OpenAlex, at import time) |
| References | 56 |
| Model or method | SVM, ANN, kNN, RF (best model not specified) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/v16010045 |
| Open access | gold |
| Keywords | machine learning, dengue, QSAR, drug repurposing, antivirals |
Importance rating
significant — thousands to millions affected
Abstract states dengue outbreaks persist in global tropical regions with no approved antivirals, necessitating therapeutic development; model aids antiviral drug discovery.