Scientists are using the same AI foundations to make drug discovery faster, safer and more effective.

Generative artificial intelligence may be having its banner moment, but the technology existed long before ChatGPT and DALL-E. It began in 2014 with a paper by Ian Goodfellow and several other researchers entitled “Generative Adversarial Networks” (GANs). Goodfellow is a computer scientist who worked for Google Brain and Apple and is currently with DeepMind. Today, his paper has been cited more than 55,000 times and underpins several AI tools.

Nearly a decade ago, Goodfellow uncovered a breakthrough: by using technology to draw on large amounts of data, AI tools can generate “synthetic” data under the right conditions. Over time, with constant training and feedback, the system learns to provide synthetic data closely aligned with the desired output. Today, these synthetic data might include smart contract code, fraud detection algorithms, and of course, hyperrealistic avatars with your face in the metaverse.

Generative AI not only solves challenges like coding and risk management but also drives powerful biotech innovations. Despite advances in manufacturing and discovery, it still takes 10-15 years and costs millions of dollars to bring a drug from discovery to market. And instead of declining with technological advances, the cost to bring a drug to market is only increasing.

AI can optimize speed and efficiency in drug discovery by streamlining new targets, designing new drugs, and even determining the likelihood of clinical trial success.

Generative AI enters the chemistry world

In 2016, Dr Alex Zhavoronkov, founder of drug discovery unicorn Insilico Medicine, made waves in the chemistry world by presenting generative AI technology at conferences from London to San Francisco. His research findings seemed farfetched to some but transformative to others–GANs, combined with reinforcement learning, could generate novel molecules for treating diseases.

Seven years ago, many still found AI a sci-fi, futuristic concept. Zhavoronkov brought examples of the technology’s ability to make something new to change people’s minds. He added petals to photographed flowers and generated unique faces to explain how AI can create new molecules. The chemists were skeptical, but Zhavoronkov was undeterred. AI was going to transform our health experiences; it just needed time.

Insilico eventually showed that its AI could find new disease targets. Using generative AI technology to produce and evaluate candidates and drug targets, their platform designed new molecules that could be synthesized, tested and developed into potential treatments.

WuXi AppTec joined Insilico to develop its first generative AI-produced molecules and later invested in the company for further acceleration. Their first drug targets may surprise you: rare disease treatments. Because these diseases are so uncommon, scientists know very little about their chemical structure. AI filled in the gaps to design potential candidates where no structure was available.

They targeted the JAK3 isoform, a DNA sequence related to rheumatoid arthritis and psoriasis. The system generated 300,000 molecules and narrowed the selection to 100 promising targets. Humans joined the process here, with medical chemists choosing the best candidate for further development. The results were published in 2018 in Molecular Pharmaceutics with a clear promise: generative AI was here to disrupt the drug discovery space.

When will AI reach our pharmacies?

Insilico secured patents on its AI technology, but it also received patents for its work on biological aging biomarkers. The company strives to leverage AI to discover powerful anti-aging treatments. While we’re several years from these reaching our medicine cabinet shelves, Insilico is also looking closely at how we age, including measuring our biological age. Aging clocks provide researchers with valuable insights into individual aging processes.

In 2020, Insilico Medicine’s generative chemistry work launched as Chemistry42. The platform uses deep learning and reinforcement learning to generate chemical structures for treating predefined medical targets. Chemistry42 identified a completely new and potentially first-of-its-kind molecule, PandaOmics, for treating fibrosis. The Insilico team designed and synthesized 80 molecules, with one small molecule showing outstanding promise for treating idiopathic pulmonary fibrosis (IPF), a rare and devastating progressive lung disease.

The company had broken new ground by uniting deep learning and chemistry. Major pharmaceutical companies noticed, too, with Pfizer, Arvinas, Fosun Pharma, and Sanofi establishing partnerships with Insilico.

By February 2022, Insilico crossed another threshold by bringing its IPF drug to Phase 1 clinical trials in under 30 months. In January 2023, those Phase 1 trials announced positive topline results, and in February 2023, the IPF drug received Orphan Drug Designation from the FDA. It’s time for Phase 2 clinical trials, where actual IPF patients will enter clinical trials and test the potentially life-changing treatment option.

Next up for AI-generated drugs? COVID-19. Insilico’s oral treatment, ISM3312, will soon enter clinical trials in China. The drug offers protection against mutations and poor outcomes for COVID patients. The world desperately needs rapid solutions to emergent diseases.

Generative AI transcends creative images and in-depth coding. It will change how physicians treat diseases and save countless lives. There’s plenty of room for blockchain too–drug discovery scientists can use DLT to securely exchange clinical research data.

My advice for the crypto community? Join the movement. Your determination brought crypto, blockchain and stablecoins to the mainstream. You can envision a future that other people can’t, and the longevity field needs your unique perspective to help bring this next generation of tech to the fore. Test AI tools in your workspace, read up on AI tokens and monitor medical news for future discoveries. We live in a unique time for technological progress, and it’s our responsibility to support its life-changing outcomes. Merging cutting-edge technology with traditional research and discovery is what will bring all of these life-changing inventions to more people.

Read More