A New Approach to Drug Repurposing with Two-Stage Prediction, Machine Learning, and Unsupervised Clustering of Gene Expression
Problem
Finding new uses for existing drugs is slow and costly; the study aimed to identify repurposing candidates by linking diseases through shared gene-expression patterns.
Why it matters
Repurposing existing drugs could speed up treatments for infectious and chronic diseases, including pandemic responses.
What the AI did
Machine learning clustered 262 disease cases and 268 controls by altered gene expression (UMAP plus k-means), then assessed drug efficacy by reversibility of abnormal gene expression to flag repurposing targets.
Results
not stated in source (abstract truncated before evaluation outcomes)
Impact
Proposes an ML workflow for drug repurposing, but demonstrated results are not visible in the provided abstract.
Record
| Venue | OMICS A Journal of Integrative Biology |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Yi Cong, Misaki Shintani, Fuga Imanari, Naoki Osada, Toshinori Endo |
| Citations | 40 (OpenAlex, at import time) |
| References | 22 |
| Model or method | Two-stage prediction with UMAP and k-means clustering |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1089/omi.2022.0026 |
| Open access | hybrid |
| Keywords | drug repurposing, machine learning, gene expression, UMAP, clustering |
Importance rating
major — millions of lives, an industry or a field
Abstract states drug repurposing has broad importance in planetary health for infectious and chronic diseases and proved important for COVID-19 pandemic interventions, implying large-scale health impact.