AIxMol® is a comprehensive technology platform that leverages artificial intelligence and computer-aided drug design (AI+CADD) to facilitate new drug discovery and development. The platform offers essential toolkits for various stages of new drug research and optimization, such as generating molecules, screening candidates, predicting ADMET properties, hopping scaffolds and estimating binding free energy. It also provides unique toolkits that cater to the specific new drug development goals of AIxplorerBio, such as designing drugs for multiple targets and predicting tissue selectivity, etc.
A platform that utilizes bioinformatics and systems biology techniques to discover dual-targets, including modules for analyzing target interactions, drug synergies, pathway interactions, and omics data.
A structure-based dual-target molecule generation algorithm that uses reinforcement learning and active learning to make dual-target molecules by joining pharmacophores and optimizing other properties at the same time.
An integrated deep learning and machine learning prediction model for the key in vivo pharmacokinetic parameters of drugs: VDss, Cl, and fu.
An artificial neural network prediction model for the tissue-plasma partition coefficient, Kp, of drugs.
A deep learning-based scaffold-hopping tool.
A scaffold decoration tool based on chemical reaction route.
A diffusion model based scaffold decoration tool.
A Graph Convolutional Network (GCN) based scoring function for protein-ligand binding affinity prediction.
A multiple-target compound recommender system.
A multi-task Graph Convolutional Network (GCN) model to predict the compound-protein interactions.
An efficient molecular docking tool combining VINA and deep learning enhanced scoring function.
Target specific molecular docking rescoring function.
An active learning assisted docking framework for Structure-based virtual screening against ultra-large compound collection.
A tool for pharmacophore modeling and compound screening.
A deep learning model, co-developed by Baidu, for ADMET properties prediction.
An integrated model of deep learning and machine learning for hERG toxicity pre-screening.
An integrated deep learning and machine learning model for oral acute toxicity pre-screening.
A tool for binding free energy calculation between small molecules and protein based on free energy perturbation theory.