Deargen Announces New Model for Optimizing Drug Candidate Molecules With Nearly Doubled Performance Compared to Existing Models
Deargen | April 12, 2021
Deargen, a drug research and development biotech company based on artificial intelligence (AI), announced on April 9 that it presented study results on controlled molecule generator (CMG) technology at the ACM Conference on Health, Inference, and Learning (ACM CHIL) 2021. CMG technology can modify many properties of molecules at the same time.
The Association for Computing Machinery (ACM), based in New York, is one of the world's most prominent academic organizations. The ACM was founded in 1947 and now has over 100,000 members worldwide. The ACM Conference on Health, Inference, and Learning 2021 was held online on April 8th as a healthcare conference.
Deargen's latest CMG technology overcomes the limitations of current models for predicting molecule properties. Deargen's test results revealed that the new CMG method enhances efficiency by nearly doubling as compared to existing versions such as MolDQN and VJTNN.
Deargen tried to improve aniracetam, which has the lowest binding affinity of dopamine D2-type receptor (DRD2) among 28 DRD2-targeted compounds in clinical or pre-clinical trials, in the most recent report. The study discovered that aniracetam's DRD2 binding affinity was highly enhanced while its other properties remained relatively unchanged.
The technology of optimizing properties of candidate molecules in the process of developing new drugs, such as efficacy, toxicity, and structural similarity, is positioning itself as a key strategy for reducing costs and time required for drug development.
Models that predict properties of previously proposed molecules, such as MolDQN, VJTNN, and VSeq2Seq, are considered to have limitations when used to produce new drugs because they either alter properties of molecules that must be retained by optimizing only one property or take very long to process data.
Deargen's CMG model, on the other hand, will optimize only desired properties while minimizing changes in other properties that are meant to be retained as set by molecule design as it can consider multiple properties at the same time. Since it takes less time to examine, it can be directly applied to the development of new drugs.
“Not only is our representative technology, MT-DTI, being used for discovering various new substances, including treatment prediction for COVID-19, but we are also poised to provide more advanced new drug development services by developing an ensemble of CMG models with our other platform MolEQ,” said Bonggeun Shin, Deargen's Chief Artificial Intelligence Officer.
Deargen President Kilsoo Kang said, "Deargen has accomplished drug development technologies on a global level and continues to invest in advancing these technologies and extending their support range." “The most recent technological advancement will provide an important momentum in the development of AI-powered new drugs.”