LeadMuta Biotechnology Files IPO Prospectus: Championing a Medicinal Chemistry-First Approach in AI-Driven Drug Discovery

Recently, a major news event occurred in the AI application field of intelligent driving: L4 autonomous driving unicorn Argo AI officially announced its closure. Argo AI had raised a cumulative $2.6 billion, backed by Ford and Volkswagen, yet struggled to achieve commercial implementation; in contrast, L2 combined driving assistance functions have been widely implemented, with adaptive cruise control, lane-keeping, and automatic parking features already becoming standard equipment in many mainstream new energy vehicles.

"Just like intelligent driving, the role of AI in the pharmaceutical field should be L2 rather than L4."AI is more of an auxiliary tool for pharmaceutical experts, andNon-substitutable pharmaceutical expert."LeadMuta's CEO, Yusu Liu, stated, Over time, the AI pharmaceuticals industry has gradually moved from hype towards rationality, slowly returning to the essence of drug development.

Unlike most AI pharmaceutical companies dominated by AI technology teams, LeadMuta,对标纳斯达克上市AI制药公司Exscientia, is an innovative drug startup focusing on ion channels and GPCRs, led by medicinal chemistry experts with AI technical personnel as support. Since its establishment in December 2021, LeadMuta has built a medicinal chemistry team mainly composed of Chinese returnee scientists. Dr. Shi Yi Yue, the founder, has nearly 30 years of experience in computer-aided drug design, having previously worked at AstraZeneca and ChemPartner.

2020 was the first year of AI pharmaceutical explosion in China, with a large amount of capital beginning to flow into this emerging track. Some companies raised 2-4 billion US dollars in Series C financing, while others completed three rounds of financing in less than a year.

The continuous iteration of artificial intelligence technology has also brought higher expectations to this industry. From machine learning algorithms to natural language processing models, from AlphaGo to AlphaFold2, people expect AI technology to revolutionize pharmaceutical research and development, bringing faster R&D speed, lower R&D costs, and higher R&D success rates.

However, it is regrettable that AI pharmaceuticals have not brought the expected results. "Many AI pharmaceutical companies in China that follow the AI platform path, despite raising substantial funds and operating for a considerable time, have shown mediocre progress and failed to deliver tangible outcomes," Liu Yusong added: "Some collaborations between domestic pharmaceutical companies and AI pharmaceutical firms have not been ideal. Essentially, pharmaceutical companies assign projects to AI pharmaceutical companies, which then return a large amount of data; however, this data offers limited assistance to the pharmaceutical companies."

The essence of AI pharmaceuticals is that computers learn and mine data to summarize and deduce patterns for optimizing drug development processes. AI possesses capabilities in learning speed and data retrieval that are difficult for humans to match.But in Liu Yusong's view, AI is more of a tool and cannot determine the outcome. AI drug development still needs to be scientist-led.

Liu Yusong believes that the main use of AI in the pharmaceutical field at present is to help medicinal chemistry experts save some time and cost, for example, by using AI technology to help them find more novel compound skeletons and improve the drug-likeness of compounds.

"People have placed overly high expectations on AI; AI in drug development cannot revolutionize the entire industry at present and is merely an auxiliary tool for medicinal chemistry experts." Currently, there is fierce competition in new drug development. If AI can help advance the drug development process by two or three months, the market value of the drug will also increase.

AI Drug Development is a Typical Interdisciplinary Field: For AI to Better Serve New Drug Research, It Must Undergo Extensive Real-World Experimental Studies. Therefore, Cross-Disciplinary Talents Who Are Proficient in Both Computer Science and Pharmaceutical Chemistry Have Become Highly Sought After in the AI Drug Development Industry. Even Leading Technology Companies Like Tencent and Baidu Have Begun to Aggressively Recruit Medical and Clinical Talents for Their AI Drug Development Subsidiaries.

However, most talents in computer-aided drug design (CADD) and artificial intelligence drug discovery (AIDD) are concentrated in foreign universities and pharmaceutical companies such as MIT, New York University, Harvard University, Merck, and AstraZeneca. "Basically, each computer-aided pharmaceutical talent can support 2 to 3 new drug development projects," Liu Yusong introduced.

In China, the innovative drug industry started relatively late, with a long-term focus on generic drugs, which almost did not require computer-aided design. Therefore, pharmaceutical professionals in China were mainly from pharmaceutical science backgrounds, and there were very few talents in computer-aided drug design in previous years.

However, pharmaceutical companies highly value whether the core technology team of an AI drug discovery company understands the pharmaceutical field and whether they can communicate and engage in dialogue with each other. Therefore, the ability to build a team with practical experience in AI + pharmaceuticals often becomes the key for AI drug discovery companies to gain a competitive edge.

LeadMuta adheres to the principle of "medicinal chemistry experts leading, AI technical personnel assisting." Dr. Shi Yi Yue, the founder and CTO of LeadMuta, has served as the Chief Scientist at AstraZeneca's Montreal R&D Center and as the Executive Director of Computational Chemistry and Bioinformatics at Pharmaron. His involvement has attracted professional talents who have long worked in multinational pharmaceutical companies to LeadMuta, forming a team of medicinal chemistry experts mainly composed of Chinese returnee scientists.

LeadMuta's computer team, relying on the Artificial Intelligence Institute of Zhejiang University, began entrepreneurship in the direction of natural language processing as early as 2015. It is one of the earliest teams in China to engage in the commercial application of AI technology engineering, with profound technical accumulation.

The value of AI pharmaceutical companies does not lie in how powerful their algorithms are, but ultimately depends on whether their projects are recognized by the industry that actually develops drugs and whether pharmaceutical companies are willing to pay for them.

“We used to talk to pharmaceutical companies about cooperation with little effect, but now when experts from the pharmaceutical and computer teams go together, it's much easier to reach a consensus."!" Liu Yusong exclaimed.

AI pharmaceutical companies are mainly divided into two categories: one is SaaS providers that mainly offer software platform services, and the other is biotech companies that develop internal pipelines.

In Liu Yusong's view, there is a severe lack of compound talents proficient in computer use within pharmaceutical companies, leading to difficulties in understanding and utilizing software provided by AI-driven pharmaceutical firms. Meanwhile, the second type of biotech enterprises struggles to compete with large pharmaceutical companies, as these established players have long-standing relationships with hospitals, drug regulatory agencies, and other institutions, making it easier for them to advance clinical research, navigate reviews, and sell products.

"Self-research + Cooperation" Becomes a New Development Model for AI Pharmaceutical Enterprises,That is, AI pharmaceutical companies advance new drug development to the PCC stage and beyond, then continue progress through collaborative research with pharmaceutical enterprises, licensing, or selling to those enterprises. Moreover, the further an AI pharmaceutical pipeline progresses, the greater its value generally becomes.

In fact, the later the stage, the more limited the scope of AI's involvement becomes. Whether a project can pass the IND review by the drug regulatory authority, and whether the animal experimental data can withstand scrutiny, primarily tests the medicinal chemistry team of the AI pharmaceutical company. The role of AI in drug development is to provide the medicinal chemistry team with directions for developing lead compound series and predict the likelihood of reaching the preclinical candidate drug standard. This often becomes the key factor for pharmaceutical companies in deciding whether they are willing to collaborate with an AI pharmaceutical company.

LeadMuta has established partnerships with tech giants like Huawei and Baidu, as well as pharmaceutical companies such as Zhonglian Pharma. Following the receipt of multi-million-dollar angel round funding, LeadMuta is now preparing for a new round of pre-A financing, primarily to advance its pipeline.

Regarding future plans, in the short term, LeadMuta "hopes to quickly advance the project to IND, achieve license conversion transactions, and validate the company's business logic." In the long term, Liu Yusong stated that it is necessary to learn from the Boston experience, establish joint laboratories with universities or pharmaceutical companies based on the company’s AI-assisted drug discovery technology, and expand more pipelines.