OBE-DB: A Computational Tool and Web Server for the Prediction of Obesity Drugs

The huge obesity prevalence and related metabolic disorders highlight the urgent need for new therapeutic strategies beyond lifestyle interventions. Despite the availability of novel pharmacological treatments, the search for more effective and safe antiobesity compounds remains a challenge. Recent advances in high-performance computational drug discovery have enabled the rapid screening and identification of potential antiobesity compounds [Correction added on 10 February 2026, after first online publication: "Methods" was deleted from this sentence.]. However, these in silico procedures frequently require complex computational knowledge that limits the use of these techniques for most researchers and hampers interdisciplinary works.

To address this gap, we have developed OBE-DB, an accessible and user-friendly platform integrating computational tools that facilitates the prediction of potential antiobesity molecules through two complementary approaches: (i) shape similarity analysis against a curated database of approved obesity drugs and (ii) inverse virtual screening of user-submitted molecules against a set of therapeutic protein targets linked to obesity.

Our results demonstrate that the server effectively screens and ranks compounds with high predicted activity, outperforming conventional in silico techniques in terms of accuracy and usability.

The OBE-DB web server represents a significant advancement by providing researchers with an intuitive tool to accelerate early-stage drug discovery for obesity treatment. The server is freely accessible without registration, providing users with a detailed report via email upon completion of the predictions. This innovative database and web server is accessible online via https://bio-hpc.ucam.edu/obe-db/.