AI4S Helps Understand Disease Mechanisms, "Zheyuan Technology" Secures Hundreds of Millions in Series A1 Round Financing | Exclusive by 36Kr

36Kr learned that Zheyuan Technology, an AI4S enterprise, recently completed a 100 - million - yuan Series A1 financing round. The investment was led by Guoke Investment (China Science and Technology Industry Investment Management Co., Ltd.), with participation from Zeyuan Fund and Ruizhi Pharmaceutical.

The domestic AI - driven drug discovery track has experienced ups and downs and is now gradually returning to rationality. With the maturity of AIDD tools such as molecular virtual screening, free - energy prediction, and antibody structure optimization, the difficulty of drug molecule design is decreasing. However, the challenges at both ends (target discovery and clinical trials) have not fundamentally changed due to technological development.

The development of molecules for mature targets has entered a highly competitive "red ocean," and the discovery of new targets is facing a shortage. Pharmaceutical companies need to invest huge amounts of funds in the long - term clinical trial stage but still face a high risk of failure. Therefore, leveraging AI technology to discover new mechanisms, evaluate new targets, and improve the efficiency and success rate of clinical trials has become an important focus in the industry.

Understand Diseases and Solve the Dilemmas at Both Ends of New Drug Development

Different from many companies focusing on "AI + molecules," Zheyuan Technology positions itself as an "AI4S + disease" company. Its "computational medicine" platform attempts to empower drug innovation with a new paradigm, especially in the discovery of new mechanisms and targets and the clinical trial stage.

Zhang Chunming, the founder and CEO of Zheyuan Technology, believes that thanks to the industry's long - term development, designing "qualified and patentable" molecules is no longer a major obstacle. The biggest challenge in current drug innovation lies in in - depth understanding of diseases.

"The first - principle of drug R & D should be to treat diseases. At the beginning of a project, systematically understanding diseases, determining the causal relationship between targets and diseases, and identifying potential indications and patient characteristics can make subsequent resource investment more worthwhile. Only in this way can the efficiency and success rate of drug R & D be fundamentally improved."

Based on the computational medicine platform, Zheyuan Technology has built a cluster of intelligent agents (Agents) for "understanding diseases," which have produced three insights in disease and target mechanism research: Discovering entirely new targets and opening up new R & D directions; discovering new mechanisms for known targets and finding differentiated indications for pipelines of the same class; and drug repositioning to expand new indications for marketed drugs, unlocking the value of old drugs, and benefiting patients.

In addition, it is reported that Zheyuan Technology's computational medicine platform has also developed the ability of "virtual clinical trials," which can be simply described as "digital humans taking digital drugs."

Its solution is "digital twin of life functions," that is, virtual patients. This is not a physical appearance simulation but maps an individual's omics data onto the human biological signaling pathway network to construct a high - dimensional mathematical model that reflects the patient's life function characteristics and disease features. In the virtual world of the computer, the results of the digital twin of the patient after being perturbed by the drug are simulated.

According to Zheyuan Technology, through virtual clinical trials, the effectiveness of a drug in tens of thousands of indication subtypes can be evaluated at the drug pipeline demonstration stage. This AI - based efficacy prediction has been verified in actual projects.

In a virtual clinical parallel trial project in cooperation with Beijing Cancer Hospital, Zheyuan Technology used its computational medicine platform to predict the drug response of eight enrolled patients. As it described, "It's like using AI to 'predict the future' for patients and forecast the drug - taking results (such as the degree of disease remission). The unblinding results showed that the AI - predicted drug response results were completely consistent with the real clinical trial results."

Facing complex diseases, using AI to predict the results of drugs in clinical trials means that before a drug officially enters human trials, suitable indications for the drug can be screened out, and trial - and - error can be carried out on the computer, accelerating the process of crossing the "valley of death" in clinical trials.

From "Ideal Utopia" to "Verifiable Results"

Currently, there are still some doubts in the market about the technological capability boundaries and commercial value of AI - driven drug discovery. Especially in complex areas involving deep interdisciplinary intersections such as new target discovery and virtual clinical trials, it is not easy to judge whether a company has these capabilities. Therefore, there is a growing trend in the industry to use results such as BD transactions or co - development contracts to prove a company's capabilities.

In Zhang Chunming's view, there is a progressive methodology for judging the ability of innovative technologies, which can be described in five levels:

"The first level is the ideal utopia, where one can see the opportunities; the second level is a unique methodology, where one can explain it clearly. A unique methodology can bring excess dividends; the third level is to establish a technical system based on the methodology; the fourth level is an engineered technical platform with reliability and scalability; the fifth level is to produce several verifiable results."

In the cross - field of AI and medicine, achieving the fifth level can prove that a team has technological innovation, maturity, and the potential to empower the industry.

In addition to the above - mentioned virtual clinical project in cooperation with Beijing Cancer Hospital, Zheyuan Technology has also produced other verifiable results. Among them, the Class 1 innovative drug PR00012 for pancreatic cancer has entered Phase I clinical trials. Through the computational medicine platform, more than 200 "insights" related to potential targets have been calculated, and each insight is expected to develop into a new drug IP asset worth more than 10 billion yuan.

Zheyuan Technology has chosen a business model of building an "IP factory" for innovative drugs. "After decades of development and the efforts of experts in various fields, the pharmaceutical industry has formed a very mature supply - chain system. Pharmaceutical factories have strong capabilities in clinical development, production, and commercialization. Our capabilities are highly complementary to all parties in the industrial chain. We hope that all parties can collaborate and innovate to efficiently transform the industry's production capacity and resources into drug assets and benefit patients," Zhang Chunming said.

By building an underlying technical system to transform drug R & D from an art into a predictable and reproducible engineering technology, and completely changing the industry's dilemma of "taking 16 years, investing $2.6 billion, and having only a 3% success rate" in new drug R & D has become the direction of efforts for many technology - innovation enterprises such as Zheyuan Technology.