News
}
AI Roundtable Discussion | Dr. Yue Qian from Viva Biotech: AI does not completely replace experiments but rather addresses challenges in each phase of drug development one by one.
Time: 2024-10-23
Source: Viva Biotech
Share:
[Abstract]:Dr. Yue Qian, the Executive Director of Computational Chemistry and Artificial Intelligence for Drug Discovery (CADD and AIDD) Platform at Viva Biotech, participated as a special guest in the roundtable discussion titled "AI in Pharmaceuticals – A Dialogue Between Pharmaceutical and TMT Analysts on AI in Drug Discovery."

Recently, Haitong Securities hosted the 2024 Shanghai Leading Industry Conference, featuring the 13th CEO Forum for the Pharmaceutical Industry and the 4th Artificial Intelligence Conference. This event brought together a distinguished group of experts, industry leaders, and representatives from investment institutions to discuss the forefront of industry trends, share the latest research findings, and explore new opportunities in leading industries, including integrated circuits, biopharmaceuticals, and artificial intelligence. Dr. Yue Qian, the Executive Director of Computational Chemistry and Artificial Intelligence for Drug Discovery (CADD and AIDD) Platform at Viva Biotech, participated as a special guest in the roundtable discussion titled "AI in Pharmaceuticals – A Dialogue Between Pharmaceutical and TMT Analysts on AI in Drug Discovery." Alongside Dr. Qian, other esteemed panelists included Other panelists included Dr. Feng Ren, CEO of Insilico Medicine; Shaochun Li, Digital Strategy Scientist at Yunnan Baiyao Group; Chang Yu, Director of Life Sciences at Amazon Web Services (AWS); Miao Luo, Head of Specialty Strategy at Sanofi (China); and the moderator, Shuqiao Yao, Chief AI Strategist at Haitong International.

 

(Dr. Yue Qian, third from the left, the Executive Director of CADD and AIDD Platform at Viva Biotech)

 

Dr. Qian introduced Viva Biotech as a one-stop drug discovery platform, emphasizing its leading advantage in structure-based drug discovery (SBDD) and its extensive experience in studying new targets, mechanisms, and molecular modalities. Discussing AI applications, she emphasized that AI tool development is centered on addressing specific challenges encountered in drug development—those which experimental and traditional computational methods have found difficult to overcome. At Viva Biotech, these AI tools are fully integrated into its existing SBDD platform and seamlessly collaborate with the wet-lab and dry-lab platforms. Dr. Qian then shared several insightful and enlightening viewpoints, drawing from her vast experience in the field.

 

Q1: How would you rate AI's current impact on drug discovery, and when do you foresee AI achieving its "ChatGPT moment" in the pharmaceutical industry?

 

Dr. Yue Qian: Firstly, we shouldn’t expect AI to replace experiments completely; rather, AI tools should focus on addressing specific issues within each stage of drug discovery. Based on the current accuracy of those tools, I would rate AI’s tools around four or five out of ten. Biopharmaceutical research is an experimental science, and model iteration and optimization require a much longer timeline compared to large language models. Should we succeed in applying computational solutions throughout the entire research process to maximize efficiency, that would be the "ChatGPT moment" for the pharmaceutical industry.

 

Q2: What are the most mature or promising applications of AI at present?

 

Dr. Yue Qian: I believe molecular generation algorithms are particularly promising, but it also faces considerable challenges. In my view, the key to breakthroughs lies in enabling models to better learn physical and chemical knowledge. This approach can also be applied to optimize other AI models. We need to move beyond generalized models and build specialized knowledge systems within this domain. However, this is by no means a simple task, given the inherent complexity of biological systems and the interactions of numerous factors. Viva Biotech's proprietary molecular generation algorithm leverages its strength in structure-based drug discovery to compensate for this shortcoming. By clarifying mechanisms of action and systematically analyzing the influencing factors, AI tools can achieve their full potential.

 

Q3: Given that the number of known drug targets is far smaller than the number of diseases, what do you consider to be the biggest challenge facing AI in drug discovery, and how can it be addressed?

 

Dr. Yue Qian: Besides the widely acknowledged challenge of data, one of the greatest difficulties lies in defining a specific problem for the model to solve. This includes understanding the underlying causes of phenomena and equipping models with specialized knowledge. At Viva Biotech, our AIDD and CADD platform, staffed by a multidisciplinary team of PhD-level experts, consistently breaks down problems, fosters cross-departmental collaboration, and encourages knowledge sharing between our computational and experimental platforms. Through persistent effort towards this goal, we are confident in achieving meaningful outcomes.

Media contact: vivapr@vivabiotech.com
Contact Us