Introduce the latest progress in the field of drug discovery, and look forward to the current status of some cutting-edge technologies and new drug discovery processes.
Wen Shuhao: AI is an inevitable trend in drug development, but we must still be cautious. Because the main problem of artificial intelligence is the generalization ability of the model, such as some new targets, or some difficult targets, or some drugs without sufficient historical data, the establishment of artificial intelligence models is very particular about When the feature vectors are different, the amount of training set data required will be different.
The PHD of theoretical physics in the company tries to understand the molecular structure and target structure of drugs in drug development from the perspective of atoms and electrons. This is our digital process.
The first step in the artificial intelligence process is to have the correct digitalization process. The lower-level digitization can capture the most essential feature vectors and make your model more generalized. Investment does not focus on artificial intelligence capabilities, but on our physical understanding of drugs.
The second angle is the research and development of artificial intelligence drugs. After all, the industry is guided by experiments. Algorithms and experiments should be connected to get feedback. Now we have started to do some automated synthesis. The inverse synthesis analysis algorithm and artificial intelligence algorithm are connected with the experimental feedback system. The game between the two algorithms may design medicines that humans cannot imagine. The chemical space is as large as 60th power. It is an inevitable trend for intelligence to show what the human brain cannot do, but it has a long way to go.
Yao Leiwen: How do we think about which part of the process from target discovery to drug marketing? Do CRO drug discovery by yourself or cooperate in development after selecting the drug. How to plan this in the future?
Wen Shuhao: Generally speaking, the drug chain is particularly long, and there will be some key links in it. Each company should do what it is better at, but a company must be bigger and it must not be limited to a very vertical direction. Jintai Technology has now moved to the early stage of medicine, and there are many subdivisions in it, including drug-making and synthesis. The next part of the crystal form is in the laboratory stage, and even the subsequent preparations. In the early business of crystal prediction, the combination of online algorithms and traditional CROs provided a holistic solution. Enterprises are very interested because this is a brand-new model, which has lower cost and faster time than the past, and solves the ultimate problem. We will do the whole process in the areas that we are good at, and even combine the new CRO mode online and offline to represent a new production mode for improving the efficiency of the pharmaceutical industry.
Concerning the business model, the most direct thing is to connect online and offline at a certain link and provide a new one-stop CRO service. There will be a good opportunity for sales volume and market share. Jintai Technology is not a pharmaceutical companyCompany, but we want to work with some real scientists to make medicine and provide the best nuggets tools. The algorithm must be the lowest level to have the opportunity to become a platform company. Once a platform company is made, there will be many unthinkable business models in the future.
This article is from the WeChat public account: Huaxing Capital (ID: iChinaRenaissance) , author: Healthcare