Tencent today launched its first AI-powered drug discovery platform, "iDrug", which will help developers to improve preclinical drug discovery.
The efficiency of discovery is expected to help the pharmaceutical industry to conduct drug development quickly and cost-effectively.
On July 9, the 2020 World Artificial Intelligence Conference Cloud Summit opened. Tencent's Chief Operating Officer Ren Yuxin announced the progress at the conference.
Tencent has entered into cooperation with a number of pharmaceutical companies to apply AI models to drug development projects.
More than ten projects, including those related to the development of drugs against novel coronaviruses, have been running steadily on the iDrug platform.
The iDrug platform aims to cover the entire process of preclinical drug development, including protein structure prediction, virtual screening, molecular design/ optimization, ADMET property prediction (soon to be open source), and synthetic route planning.
Protein structure prediction, as the basis of drug design, is essential for understanding the interactions between molecules in an organism.
Previously, pharmaceutical companies and research institutions have conducted experimental determination of protein structure through traditional methods, which are often difficult, lengthy, and costly.
Whereas, after predicting protein structure and function through deep learning models, computers can faster from hundreds of millions of massive small molecules. Quickly and specifically find potential seedling compounds to effectively improve R&D efficiency.
On the iDrug platform, Tencent AI Lab has applied a new algorithm for predicting protein structure.
The data shows that Tencent's new algorithm has a significant improvement in hard cases, which is better than the industry-recognized authoritative method Robetta Improved by 10%.
Since joining CAMEO in 2020, the authoritative global testing platform for protein structure prediction, the Tencent AI Lab team has With this self-developed algorithm, it has won the monthly title five times in six months.
The innovative ideas of this algorithm have also been applied to the iDrug platform, which will be further leveraged for new target discovery and disease mechanism research.