“AI Pharma” is launched in Chengdu. Why is it regarded as a “game-breaking artifact” for pharmaceutical research and development?

A little girl with leukemia lay in a white medical cabin. The scanner in the cabin scanned the whole body of the little girl. The screen quickly showed that “100% of cancer cells have been removed”, and she was cured—— This is an excerpt from a sci-fi movie.

Humans can visualize the future of AI medical treatment. Intelligent equipment and systems such as Da Vinci surgical robots and “AI + medical imaging” have long been widely used in the clinical field, but AI pharmacy (AIDD, AI drug discovery) is covered with a layer of A veil of mystery.

▲Screenshot of the medical cabin movie in the sci-fi movie “Elysium”

AI pharmaceuticals, that is, machines can empower different fields of drug research and development through learning and mining data. Right now, what the layman sees is usually just an obscure string of letters and numbers in a large computing device. But in the eyes of professionals, it is becoming a “game-breaking artifact” in pharmaceutical research and development.

According to reports released by a number of think tanks and research institutions, in 2020, China’s AI pharmaceutical industry will usher in explosive growth. Giants such as Tencent, Alibaba, and Baidu have entered the market one after another, and a large number of AI pharmaceutical companies have been established and obtained financing. From the perspective of industry insiders, the powerful algorithms and computing power of AI technology can currently be empowered in the two links of drug discovery and preclinical research, which play a positive role in improving R&D efficiency, reducing R&D costs, and avoiding R&D risks. 

According to media reports, in the domestic AI pharmaceutical track, the Yangtze River Delta has become a development highland, and the Greater Bay Area is also accelerating its layout. In February this year, Peide Bio, an AI pharmaceutical company born in the Yangtze River Delta, moved westward to Chengdu and obtained financing. Since then, the name of Chengdu has been on the map of AI Pharma. Recently, a reporter from Red Star News walked into the company’s R&D base and chatted with members of its team of scientists about AI Pharmaceuticals in Chengdu.


Reduce development time and costs

Mainly used in drug discovery and preclinical research

In a computer room of Peide Biomedical Co., Ltd. in Tianfu International Biological City, Chengdu, an AI device called “Peptide Server” continued to beep, and a series of codes were written on the screen, and a group of numbers were constantly being written on the screen. Variety. This AI device usually does not require human operation, but relies on programs written by engineers to actively learn and calculate. When the Red Star News reporter came to this computer room, several R&D personnel were discussing the temperature, humidity and network environment in the computer room.

Next to the computer room are several laboratories, and R&D personnel in white coats focus on the work at hand. Currently, they are working on extracting active ingredients from traditional Chinese medicine. In the future, the results calculated by the AI ​​equipment in that computer room may be able to be applied in the laboratory.

▲A running peptide server in the computer room

According to the relevant report released by the qubit think tank, AI pharmacy refers to the combination of AI technologies such as NLP, deep neural network, generative model, etc. with traditional pharmacy links, through data cross-comparison, accelerated screening, de novo generation, etc., to improve the research and development of new drugs Efficiency and technological application to expand the space for drug innovation. Technically, it will also involve computational biology, computational chemistry, computational physics and other aspects. At present, AI pharmaceuticals mainly play a role in drug discovery and preclinical research.

Rong Mingqiang, chairman of Peide Bio, told reporters that AI technology has been applied to a pulmonary fibrosis drug that the company is developing. “Pulmonary fibrosis is a clinically difficult disease to treat. At present, the drug we have developed is about 100 times more effective than existing drugs in animal experiments. This is a case that we have done well at present, and AI Technology in it helps us to provide us with the direction of transformation when we do drug transformation.”

Rong Mingqiang introduced that drug development (from initial research to launch) usually takes ten years and costs billions of dollars. Therefore, every pharmaceutical company is looking for ways to accelerate development and reduce costs, and AI technology is a better solution. Program. “AI technology can greatly shorten drug development time and reduce drug development costs. In the pharmaceutical field, time and cost are two crucial factors.”

▲Rong Mingqiang

The company’s main project is the research and development of peptide drugs derived from animal toxins. The successful cases of “fighting poison with poison” in history depend on probability and luck, and there is no systematic modern scientific theory to summarize, so scientists have been inspired to find the answer.

“For example, spiders bite their teeth and scorpions have venomous glands on their tails, which are rich in polypeptide molecules. These polypeptides have strong functions and are a molecular treasure trove for drug development.” A researcher of the company introduced that spiders are used in traditional Chinese medicine. , scorpions, snakes, etc., and the team’s work is to modernize traditional Chinese medicine, with the goal of building the world’s largest natural active peptide sample library, “and AI technology will help us discover and screen peptide molecules, and will also use AI Expand the virtual library of peptides.”


Need to understand both pharmacy and AI

Fancy development space choose Chengdu

“The research and development work is very intense now. We have been in Chengdu for 8 months, and we seldom ‘go into the city’ to play.” A research and development staff of Peide Biology told reporters. Tianfu Biocity, where the company is located, is located in the south of Chengdu, 4 kilometers away from the nearest subway station. Many people come and go here wearing white coats and walking around with a cup of coffee in hand, which is different from the old city of Chengdu. The laid-back atmosphere.

In 2016, Chengdu Tianfu International Bio-City with a planned area of ​​44 square kilometers was established, focusing on five sub-sectors of biotech drugs, innovative chemical preparations, high-performance medical devices, biological services and general health services, dedicated to building the development of the domestic biopharmaceutical industry New Heights. According to the introduction of the Biological Industry Bureau of Chengdu High-tech Zone, as of June this year, a total of 180 projects have been settled, with a total investment of over 120 billion yuan, 5 Nobel Prize teams, 4 national academician teams, and 51 high-level talent teams. With more than 7,000 talents, 5 companies have been promoted to go public, and 109 functional platforms of various types have been introduced.

Founded in May 2021, Peide Bio moved westward from the Yangtze River Delta to Chengdu in February this year. In the first half of this year, the company completed an angel round of financing of tens of millions of yuan. This round of investment was led by Chengdu Biocity No. 1 Equity Investment Fund.

Its chief scientist team has more than 20 years of experience in peptide research and development. Among them, Mingqiang Rong is a “Xiaoxiang Distinguished Professor” of the National and Local Joint Engineering Laboratory of Animal Peptide Drug Innovation and Intelligence of Hunan Normal University, a doctoral tutor, a visiting scholar at Baylor College of Medicine, and a senior returnee; chief scientist Lai Ren, a 2010 national The winner of the Outstanding Youth Fund has academic achievements such as the discovery of the world’s largest number of antimicrobial peptides (more than 500 species); Professor Gu Douxin, a researcher-level senior engineer who was the research and development director of Jiuzhitang, also joined the Peder Biological team.

“We understand pharmaceuticals and AI, but we are relatively short in the field of AI.” The relevant person in charge of the company said that the technical threshold for AI pharmaceuticals is very high, and it is necessary to understand both pharmaceuticals and AI. Therefore, its team has also absorbed well-known companies. Zhang Long, an AI technology expert. After coming to Chengdu, their accumulated knowledge and experience will be deeply integrated with Chengdu’s talents, technology, equipment and other supporting facilities.

In Rong Mingqiang’s view, Sichuan University, University of Electronic Science and Technology and other colleges and universities provide talent sources for their teams, and Chengdu’s technology and equipment industry chain is relatively complete. “We don’t care much about whether it is in the Yangtze River Delta or Chengdu, because both places meet our needs. We hope to be a long-term enterprise, and we care more about the future development space. We think that in a few decades, Chengdu will be The development is immeasurable.”

▲Researchers in the laboratory


A number of Internet giants have entered the game

Opportunities and challenges coexist

In 2012, a British AI pharmaceutical company named “Exscientia” was established, and AI pharmaceuticals began to sprout around the world. Reports released by a number of think tanks and research institutions pointed out that compared with the long and expensive law of “ten years plus one billion” for traditional pharmaceuticals, AI pharmaceuticals that “save time and money” are releasing huge potential.

It is reported that in the first quarter of 2022, more than 40 financings were initiated in the field of AI pharmaceuticals, with a total amount exceeding US$2.4 billion. Investment and financing activities still mainly occurred in China and the United States, accounting for 80% of the total financing events. In the second quarter of 2022, there were 25 investment activities in the field of AI drug research and development, with a total amount of about 8 billion yuan (incomplete statistics). According to data from Deep Pharma Intelligence, as of the first quarter of 2022, there are more than 56 pharmaceutical giants, more than 495 AI pharmaceutical companies, and more than 31 leading Internet companies in the world participating in AI drug research and development. The AI ​​pharma track is crowded with pharmaceutical giants such as GlaxoSmithKline, Merck & Co, Pfizer, etc.

2020 is known as the first year of China’s AI pharmacy. According to incomplete statistics from Bingjian Science and Technology Research Institute, before 2020, there were less than 10 AI pharmaceutical companies in the country. By 2021, at least 60 domestic AI pharmaceutical companies will be established or pharmaceutical companies will expand their business into the field of AI pharmaceuticals. Ten AI pharmaceutical companies have obtained angel round or A round of financing, and several companies have received B round or above round of financing, and the industry has begun to experience explosive growth. In the whole year of 2020, the investment and financing of domestic AI pharmaceuticals exceeded 3.1 billion yuan, an increase of nearly 7 times year-on-year. In 2021, it will reach a scale of more than 10 billion. Among them, Internet giants such as Tencent, Alibaba, Huawei, Baidu, and ByteDance have entered the game.

The “AI Pharmaceutical In-depth Industry Report” released by the qubit think tank believes that the AI ​​pharmaceutical industry is gradually moving towards a short-term plateau after rapid growth, and is expected to usher in the next round, and even the third round of explosive growth, which will depend on 2023. Two key time points are ~2024 (batch drugs complete clinical phase II) and around 2026 (appearance of AI-validated drugs on the market). It is estimated that by 2025, the total domestic market size of AI pharmaceuticals will be about 7.2 billion yuan. By 2035, the market size of self-developed pipelines will be about 68 billion yuan, and the market size of foreign cooperation pipeline research and development will be about 135 billion US dollars. Adding other related income, the total domestic market size will reach 204 billion yuan.

AI pharma also faces serious challenges. According to the data from the AI ​​pharmacy think tank “Intelligent Pharmacy”, as of June 20, 2022, there are 26 companies in the world, and about 51 drug pipelines that are assisted by AI to enter clinical phase I. However, so far, no drug developed mainly by AI has been approved for marketing.

Wang Shiqiang, a senior researcher at Bingjian Science and Technology Research Institute, believes that in the long run, the application of AI technology to pharmaceuticals will be an inevitable trend, and all pharmaceutical companies will eventually use artificial intelligence technology to promote drug R&D and production. However, in the short term, AI pharmaceutical companies face many insurmountable life-and-death barriers, especially for companies that have obtained C-round financing and above, investors’ expectations for the launch of their innovative drugs have become a huge pressure for AI pharmaceutical companies. Although some AI pharmaceutical companies’ innovative drugs have entered the clinical trial stage, it will take 5 to 10 years before they are approved for marketing. Coupled with a 90% failure rate, the prospects are still not optimistic.

How to make profits from AI pharmaceuticals is also a difficult problem on the face. “This is a core issue. At present, it is still difficult for AI pharmaceuticals to make large-scale profits.” Rong Mingqiang said that the commercialization of AI pharmaceuticals has two directions. One is to be a service provider, but it is limited by the maturity of technology. Other problems, the number of customers is limited; the second is to do drug research and development, but there is no commercial practice for any drug that can be marketed.

The reporter noticed that although AI pharmacy is still in its infancy and faces many challenges, observers and the industry generally affirm its long-term value and regard it as an inevitable trend. Rong Mingqiang said: “In the next ten years, AI will definitely become an important means of drug discovery.”

Red Star News reporter Wang Tuo

Edited by Yixi Chen