Recently, the German pharmaceutical giant Merck Group announced that it has reached a cooperation with Insilico Medicine, an AI medicine research and development company, to integrate a new molecular design product Chemistry42 generative chemistry AI platform into the Merck Group’s high-performance computing (HPC) infrastructure. Provide customized services.
Chemistry42 is the core part of the Pharma.ai drug discovery platform under Insilico Medicine. It combines artificial intelligence and machine learning methods with expertise in the fields of medicine and computational chemistry to design new small molecules with specific physical and chemical properties. The Chemistry42 platform is a scalable distributed network application that can complete multiple tasks simultaneously within a few hours. Through container orchestration and workflow management, the platform can achieve cross-hardware predictive resource allocation, and services can be provided on the cloud or local HPC infrastructure.
"AI has great potential in changing the drug discovery process, as evidenced by the flagship product of this generative chemistry AI platform," said Joern-Peter Halle, global research leader of Merck Group's medical business unit.
The German Merck Group is a pharmaceutical giant with a history of more than 350 years. It is mainly engaged in the pharmaceutical, life science and chemical business. It develops innovative prescription drugs, over-the-counter drugs, life science solutions, and industrial effect pigments. And chemicals.
Insilico Medicine is a leading AI drug research and development company, which focuses on target discovery, small molecule generation and clinical trial result prediction, using deep generative reinforcement learning technology. Since 2015, Insilico Medicine has taken the lead in the world to use generative confrontation networks (GANs) and reinforcement learning (RL) to design new drug molecules, published more than 100 research papers in the field of drug discovery, and applied for more than 25 patents. Including proof-of-concept research and experimental verification. This year, the company launched the Chemistry42 generative chemistry platform and deployed services within the first batch of large pharmaceutical companies and drug discovery partner institutions.
Since Ian Goodfellow published an original paper on Generative Adversarial Networks (GANs) in 2014, Insilico Medicine has begun to develop generative chemistry and generative biological algorithms. In 2016, Insilico Medicine published the first related paper describing the application of generative adversarial networks in small molecule drug discovery in the field of oncology. From 2016 to 2020, Insilico Medicine published more than 40 papers and obtained multiple patents in this field. Insilico Medicine has carried out a number of proof-of-concept experiments to prove that the generative model can indeed discover new targets, and design molecules with specific properties that can be synthesized and tested in vitro and in vivo.
This is a phenomenon worth paying attention to for players on the track. It can be seen that the head company has entered the harvest period, and the number of pharmaceutical company cooperation has begun to increase. This is because, for the AI new drug development track, high financing does not bring enough confidence to the industry. What companies and investors want to see more is the verification of platform AI capabilities. The verification method can be paid by the pharmaceutical company, or it can be a positive clinical result.
The industry has entered the competition among top players. several Chinese local leading companies have also completed new financing, and the funds are used to advance the research and development pipeline. The news has not been made public.
Start-ups in the medical artificial intelligence field are concentrated in medical imaging, auxiliary diagnosis and disease risk prediction subdivision tracks. There are 154 artificial intelligence projects in the medical field, of which more than 70% are in the angel round and A round stages.