Insilico Medicine Introduces Groundbreaking AI-Generated and AI-Discovered Drug in Latest Research Publication

Insilico Medicine, a biotech startup with headquarters in Hong Kong and New York, has raised over $400 million to integrate biology, chemistry, and clinical trial analysis through advanced AI systems. The company announced a significant milestone today with the release of a new paper in Nature Biotechnology, detailing the journey of INS018_055—the first AI-generated and AI-discovered drug currently in Phase II clinical trials.

The paper outlines the development of INS018_055, a drug candidate intended for treating idiopathic pulmonary fibrosis, a rare and aggressive lung disease. It presents raw experimental data along with the preclinical and clinical evaluations of this potentially first-in-class TNIK inhibitor, conceived and designed through generative AI techniques.

According to Insilico’s founder and CEO, Alex Zhavoronkov, INS018_055 is groundbreaking as it was the first drug whose target was identified using biology AI and whose molecule was created with generative chemistry AI. The company utilized its Pharma.AI platform, featuring multiple AI models trained on millions of data samples for various tasks. Key tools in this platform include PandaOmics, which swiftly identifies and prioritizes disease-relevant targets, and Chemistry42, employing deep learning to rapidly design new drug compounds targeting the identified proteins.

Zhavoronkov mentioned, “When we published our first paper on generative AI for novel molecule generation in 2016, the drug discovery community was skeptical. The progress of INS018_055 serves as proof-of-concept for Pharma.AI and demonstrates the immense potential of generative AI to accelerate drug discovery.”

Drug development is typically a lengthy and risky endeavor, requiring decades of preclinical and clinical trials, with costs often exceeding billions and a failure rate above 90%. Insilico estimates that with traditional methods, developing INS018_055 would have taken over $400 million and six years. However, using generative AI, the company expedited the first phase of clinical trials to just two and a half years at a fraction of the cost.

Kai-Fu Lee, chairman and CEO of Sinovation Ventures, an investor in Insilico, stated that the company's approach represents a “breakthrough in efficiently discovering medicines from scratch by leveraging massive data processing capabilities through generative AI in chemistry and biology.”

Zhavoronkov expressed cautious optimism about their achievement, marking it as a significant milestone in AI-driven drug discovery—a decade of work since founding Insilico in 2014. “I’ll go and watch Dune,” he remarked, “with a big bag of popcorn.”