Developing new medicines and drugs is a laborious and timely process, especially since determining how a new drug might interact with the body’s molecules and proteins often involves trial and error. Before a potential candidate drug is discovered, hundreds before it have likely been discarded–and even then, proceeding to clinical animal and human trials with the candidate drug often results in failure as well. This is one reason new designer drugs are so costly, with Big Pharma spending billions on R&D. However, some new companies are hoping to change that, or at least speed up the process. Several AI drug discovery companies have formed in recent years, hoping to leverage the power of generative AI. If this AI-power drug discovery works out, then major changes are ahead for both the pharmaceutical industry and society.
(Would you work for an AI boss? Read this Bold story and decide.)
Notably, there are different approaches to drug development using AI-powered drug discovery. Some prefer utilizing massive amounts of biochemical data to help refine AI processes in drug research. Others believe refining AI first is better, avoiding the need to acquire large amounts of biochemical information. In either case, however, AI drug discovery companies could cut the process in half if not more. That of course depends on how much better AI is in designing drugs that are safe and effective. The industry is still very young in its efforts, but the potential it offers is profound. Big changes are around the corner, but the impact of these changes remains to be seen.
A Peek at AI-Powered Drug Discovery
Previously, collecting data from various biochemical reactions in the lab was painstakingly tedious. Lab technicians and researchers would be responsible of data collection and analysis. That’s no longer the case, especially when it comes to AI drug discovery companies. Giant data labs are being created that are nearly fully automated. Not only are robotics responsible for collections and transfer. But chemical wells are made with silicon chips that can record and monitor millions of chemical reactions. To put this in perspective, a large AI data-focused lab can readily collect 50 terabytes of raw data per day. That’s the equivalent of about 12,000 full-length movies! This is the promise of AI-powered drug discovery.
The data collected at these AI drug discovery companies involves how proposed proteins interact with chemicals at a molecular level. Based on billions of recorded reactions, data patterns are assessed using AI to digitally design a new potential drug molecule. This molecule is then translated into a physical molecule using 3D printing and high-speed automated labs. This molecule can then be tested in the lab and further refined with AI-powered drug discovery techniques. Ultimately, this process, which is hopefully much faster than analog approaches, yields a stronger candidate for treatments. This greatly accelerates the drug research and development process in theory.
Shortening and Improving the Pipeline
Typically, a new pharmaceutical medication can take anywhere from 10 to 15 years to develop. About half of this time is spend in the lab before being ready for clinical trials. The remainder of the time is then spent going through phased trials with the hope a drug will be FDA-approved. This process generally costs as much as $1 billion for a single successful drug. This is not only due to the length of time of development but also due the failure rate. Of all the drugs that make it to clinical trials, over 90% fail due to lack of effectiveness or the presence of side effects. Understandably, any talent that shortens this time delay and improves success rate attracts attention from Big Pharma. And AI-powered drug discovery is such a process. This is why Big Pharma is investing in smaller AI drug discovery companies.
Though definitive numbers are not known, some expect AI-powered drug discovery techniques to markedly speed up things. Rather than the preclinical portion taking up to seven years, this could be cut in half. And if AI turns out to be better at identifying drug candidates, the success rate during clinical trials could also be improved. For AI drug discovery companies, improving the process offers some major advantages. For one, Big Pharma is willing to pay hundreds of millions for proof-of-concept. And if an AI-designed drug becomes approved for use, major royalty payments would follow. With this much money being pumped into the pharmaceutical R&D arena, several AI drug discovery companies have emerged. And each appears to be taking different approaches.
The AI Drug Discovery Company Sector
One might think that Big Pharma may lead the field among AI drug discovery companies. But in reality, there are a number of smaller-scale players in this sector investing in AI-powered drug discovery. Terray Therapeutics is among this field with its own Terra chip that has 32 million micro-wells for testing biochemical reactions. Other well-known companies in this industry include Recursion Pharmaceuticals, Shrodinger, and Isomorphic Labs. Notably, Isomorphic Labs is supported by Google DeepMind and uses an AI platform named AlphaFold 3. Each of these companies are receiving supportive venture capital for their efforts from Big Pharma or elsewhere. Despite evidence that AI-powered drug discovery works, the assumption among most is that it will.
Interestingly, companies like Terray Therapeutics and others believe the more data the better. As a result, their process for AI-powered drug discovery focuses on generating as much data as possible. In contrast, Isomorphic Labs believes better AI software systems is the answer. If AI systems can better predict biochemical interactions from the start, less actual biochemical data is needed. Both pursuit strategies have merit at this point in time. But it may be that one or the other is more successful in their approach and outcomes. In either case, pharmaceutical R&D has definitely entered into a new phase. The industry’s future looks to be at least partially in the hands of AI drug discovery companies. The next decade should therefore be quite intriguing in terms of pharmaceutical innovations.
