DBPP-Predictor provided a valuable tool for the prediction of chemical drug-likeness, helping to identify appropriate drug candidates for further development.

Welcome to our advanced platform that leverages the multitask-unimol model and rdkit to predict drug druggability through SMILES strings with comprehensive analysis of ADMET properties.

Bilingua-DPP-IV is an online tool that utilizes advanced machine learning models to predict whether peptide sequences exhibit Dipeptidyl Peptidase-IV (DPP-IV) inhibitory activity. DPP-IV inhibitors show significant potential in treating metabolic diseases such as type 2 diabetes. This tool aims to provide researchers with a rapid and convenient platform for screening potential DPP-IV inhibitory peptides. We employ a dual-modal fusion deep learning model, integrating sequence information and molecular characteristics for prediction, aiming for high accuracy and reliability.

A comprehensive source and free tool for evaluating chemical ADMET properties

admetSAR provides the latest and most comprehensive manually curated data for diverse chemicals associated with known Absorption, Distribution, Metabolism, Excretion and Toxicity profiles. admetSAR created a user-friendly interface to search for ADMET properties profiling by name, CASRN and similarity search. In addition, admetSAR can predict about 50 ADMET endpoints by our recently development chemoinformatics-based toolbox, entitled ADMET-Simulator which integrates high quality and predictive QSAR models. admetSAR will be helpful for in silico screening ADMET profiles of drug candidates and environmental chemicals.

MetaADEDB is an online database we developed to integrate comprehensive information of adverse drug events (ADEs).

NetInfer is a web server for prediction of targets and therapeutic and adverse effects via network-based inference methods. Compared with our previously developed standalone version of NetInfer, this web server provides a user-friendly interface. With the web server, users can easily predict potential target proteins, microRNAs, pathways, Anatomical Therapeutic Chemical (ATC) classification codes or adverse drug events (ADEs) for small molecules of their interests in a few steps. The prediction results may facilitate the discovery of new therapeutic and adverse effects for the user-submitted molecules, and help explain their molecular mechanisms.

Optimize your lead compound with consideration of ADMET properties.

web server for in silico prediction of sites of metabolism for drug-like compounds

A web server for metabolic prediction of human aldehyde oxidase

This repository contains the implementation of the paper CLaSP: A Contrastive Learning-Guided Latent Scoring Platform for Comprehensive Drug-Likeness Evaluation.

Bilingua-DPP-IV is an online tool that uses advanced machine learning models to predict whether a peptide sequence has dipeptidyl peptidase IV (DPP-IV) inhibitory activity. This code project is the training and testing code of Bilingua-DPP-IV.

ERRalpha-Predictor