Copyright forbes
Will AI solve the challenges of drug development? Will AI transform the typical trials and tribulations of drug development? Developing a new drug costs more than $2 billion and takes more than 10 years, but 90% of drug candidates fail in clinical trials. Data from multiple sources, well-trained machine learning and AI models, and human expertise—both in AI and drug development—could tip the scales. Developing a specialized AI platform and testing and refining it over many years is a key success factor, but no less critical is integrating it with drug development expertise and human intelligence. “You need to understand why and how and what would make a good drug target and what would make a good drug out of it,” says Anat Cohen-Dayag, Executive Chairman of Compugen. Anat Cohen-Dayag, PhD, Executive Chairman, Compugen 2025 could mark a tipping point in the application of AI to the life sciences, accelerating the discovery of novel therapeutics and enhancing the efficiency and efficacy of the drug development lifecycle. This has become a key investment theme for established pharmaceutical companies, startups, and AI leaders. Recent examples include Nvidia supporting Ely Lilly’s “AI factory” that will train “biomedical foundation and frontier models" for drug discovery and development; Isomorphic Labs, which “reimagines the entire drug discovery process from first principles with an AI-first approach,” has raised $600 million in series A funding round last March; drug discovery engines accounted for most of the $3.8 billion of AI funding in drug R&D in 2024. MORE FOR YOU Overall, AI is projected to generate over $350 billion in annual value for the pharmaceutical sector, according to CB Insights. These types of predictions, however, could be a sign of an AI bubble, where many startups describe themselves as AI developers, and investors and executives assume that lots of data and vast computing power are all that is needed to find a new drug target. Compugen was founded in 1993 when “AI” was called “computational discovery.” Initially, it provided pharmaceutical companies with hardware—a fast processor for genome sequencing—and later transitioned to offering software and data. After earning a Ph.D. in Cellular Biology from the Weizmann Institute of Science, Cohen-Dayag joined Compugen and, in 2010, became the company’s CEO. Pharma executives at the time were frustrated with the failure of genome sequencing to deliver on its expected results—finding a cure for many diseases—and Cohen-Dayag decided to exit the data provision service business. “There was no way I could convince pharma to enter into collaborations with us based just on our computational capabilities,” says Cohen-Dayag. “We wanted to be in control, we wanted to be in the driver's seat and to retain more value.” Compugen then went through a remarkable journey of transformation, going up the value chain of their industry, “from being a computational service provider to being a clinical stage innovator, discovering its own product candidate and feeding its own pipeline,” says Cohen-Dayag. Compugen’s proprietary AI-based predictive discovery platform has led to the identification of novel drug targets and the establishment of an innovative and differentiated immuno-oncology pipeline. The company has established strategic partnerships with large pharma companies, including Bayer, Bristol Myers Squibb, Gilead, and AstraZeneca. Compugen’s current “four clinical stage programs serve as a product opportunity, but also as a proof of concept for our computational capabilities,” says Cohen-Dayag. These capabilities have been honed over many years of trial and error: “We have failed many times, but we've learned how to feed back the learning from failures” into the continuous refinements of Compugen’s AI drug discovery and development platform. In September, Cohen-Dayag assumed the newly created role of Executive Chair of the Board of Directors at Compugen, and Eran Ophir, who has been with the company since 2015 and has served as Chief Scientific Officer for the past three years, was appointed President and Chief Executive Officer. Editorial StandardsReprints & Permissions
