During an online meeting, Alex Zhavoronkov, co-founder and CEO of Insilico Medicine, showed me a small green diamond-shaped pill. This pill belongs to his company, and it is to be used on idiopathic pulmonary fibrosis, a rare lung disease that has no known cure. The FDA has not approved this drug; however, it is very effective based on the initial pilot trials. The emergence of this drug is a significant accomplishment for a company in AI Drug Discovery, in which artificial intelligence is heavily involved in finding treatments.
‘We cannot state that we have the first molecule approved, identified, and designed by artificial intelligence,’ Dr. Zhavoronkov notes. Of course, one may dispute this and say that other countries are more advanced. Hence, they have wider roads. But we may be the furthest along the path.
What is AI’s Role in Transforming Drug Discovery?
This is not just another evolution because AI Drug Discovery technology is fast replacing itself as the key driver in the discovery of new medicines. Large pharmaceutical firms, as well as new biotech startups, are now in the process of adopting AI as a means of speeding up drug development.
One of these players is Alphabet, the parent company of Google, which began UK-based Isomorphic Labs in 2021. Its CEO, Demis Hassabis, recently claimed a Nobel Prize in Chemistry for an AI model that could greatly help with drug design.
Leaning on AI Drug Discovery has pros. THE CONCEPT OF AI DRUG DISCOVERY TO ENHANCE PATIENT CARE. The time it takes to develop a new drug can be anywhere from 10 to 15 years and at least $2.2 billion. The reasoning is that AI could reduce the time taken and the costs more than the success rates would improve.
Charlotte Deane, a professor of structural bioinformatics at Oxford University, supports this. ‘’We are still just scratching the surface of how good this could be,” she said, obviously underlining artificial intelligence’s growth.
How Can AI Accelerate the Path to Market?
Pharmaceutical development continues to be expensive and highly risky. The drug dropout rate in clinical trials was previously around 90%. When AI Drug Discovery is used, scientists hope to reduce this risk and expedite the course to successful treatments.
New studies show that at least 75 “AI-discovered molecules” have progressed to clinical trials, and many more are coming. However, the real mark will be made when these drugs are successfully tested and have hit the market.
Most significantly, there was no defined consensus regarding the interpretation of what constitutes an ‘AI-discovered’ drug. ‘In every case so far’, Professor Deane underscores ‘there has still been substantial human involvement’.
AI is mainly applied in two of the stages of drug development.
Identifying the Therapeutic Target:
AI is similar to genomic or transcriptomic data in that it phases out the associations with molecular signatures like genes, proteins, etc. In the past, this work only required testing on laboratory apparatus. At the same time, A, I can quickly link molecular biology and diseases.
Designing the Drug:
Generative AI, similar to the platform underlying ChatGPT, envisioned that the molecules could bind to the target. This displaces the labour-intensive and expensive system where chemists produce hundreds of different molecular forms to identify the best formula.
What Are Insilico Medicine’s Major Achievements?
Launched in 2014, Insilico Medicine has raised more than $425m and is the first and only AI Drug Discovery company to operate in all stages of development. The firm employs generative AI software that constructs drugs from scratch through target recognition, molecule design, and clinical trial prognosis.
“Machines dream until they find a perfect drug that meets all our needs,” said Dr. Zhavoronkov.
Another impressive accomplishment of Insilico is a new molecule that can help address IPF. The company’s AI found that TNIK, a protein not discussed in this disease before, was the central controller. With the help of generative AI, Insilico decided to use a molecule to block TNIK. A process usually requires four years, and 500 synthesized compound types were achieved in 18 months, but only 79 compound types have been synthesized.
While Insilico’s IPF drug is still a drug candidate, it has six molecules in clinical trials, four more set to go through trials shortly, and almost thirty other molecules with potential.
What Challenges and Opportunities Exist in AI Drug Discovery?
However, new challenges persist even as developed countries trump the rest of the world on several fronts. A major challenge is the small amount of information available for AI systems to train on. This limitation impacts both target identification and molecule design and can cause biases.
Recursion Pharmaceuticals, a US company, is on the right track, given its approach of performing automated experiments to create vast amounts of data. It then teaches its AI tools to analyze this data and look for patterns it has not caught on.
Recursion has procured the most advanced supercomputer of any pharma firm to complement its drive. Of the four molecules it is working on, one has been developed to treat lymphoma and solid tumours and is awaiting the result of a phase-I −clinical trial.
“Our AI then recognized a way of hitting a gene that has proven to be an insurmountable challenge until now,” Recursion CEO Chris Gibson also said.
What Does the Future Hold for AI-Driven Drug Discovery?
While AI-discovered drugs are entering clinical trials at an unprecedented rate, their ultimate test will be their ability to succeed in these trials.
“What matters most is proving that AI can deliver a higher probability of success than traditional methods,” says Gibson. “When that happens, it’ll be obvious to the world that this is the way to go.”
Integrating AI into drug discovery represents a paradigm shift for the pharmaceutical industry. Dr. Zhavoronkov says, “The journey is just beginning, but the potential is limitless.”
Conclusion
With advancements in AI Drug Discovery, the future of medicine looks brighter than ever. Whether through Insilico Medicine’s innovative IPF treatments or Recursion Pharmaceuticals’ groundbreaking cancer therapies, AI is poised to transform how we combat disease.
As Professor Deane aptly summarizes: “We are standing at the dawn of a new era, and the possibilities are just starting to unfold.”
Leave A Comment