The Next Era of AI Drug Discovery? Quantum Computing

A consortium of international scientists has taken the first step toward incorporating quantum computing into artificial intelligence (AI)-powered drug discovery. Already, generative AI is being used to dream up new molecules that act on new targets in order to treat diseases. With the additional processing power of quantum computing, scientists hope that they can further accelerate and refine the process.  

“Quantum computing is recognized as the next technology breakthrough which will make a great impact, and the pharmaceutical industry is believed to be among the first wave of industries benefiting from the advancement,” said Jimmy Yen-Chu Lin, PhD, GM of Insilico Medicine Taiwan and corresponding author of a new paper in the Journal of Chemical Information and Modeling that combines quantum computing and generative AI for small molecule design.

The team was led by scientists from Insilico Medicine, a clinical stage AI drug discovery company and a pioneer in the use of generative adversarial models (GANs) to design new molecules, as well from the University of Toronto’s Acceleration Consortium and the Hon Hai (Foxconn) Research Institute in Taiwan. 

In the traditional GAN model, there is a generator and a discriminator. The generator takes random noises, creates a molecule, and the discriminator attempts to distinguish between real and fake samples. This training process takes place until the discriminator can no longer tell the difference between the generated data and the real data. GANs have proven incredibly successful in AI drug design and are used as part of Insilico Medicine’s Pharma.AI platform which processes massive quantities of data to identify new targets for disease and design new molecules and has produced two drugs currently in clinical trials and 31 in its pipeline for diseases including fibrosis, cancer, and COVID-19. 

In this new paper, researchers substituted distinct parts of the GAN model with a variational quantum circuit (VQC) and compared its performance with the classical counterpart. 

The study not only demonstrated that the trained quantum GANs can generate training-set-like molecules, but that the quantum generator outperforms the classical GAN in the drug properties of generated compounds and the goal-directed benchmark. And the quantum GANs can generate valid molecules and outperform traditional GANs with far fewer input parameters.  

“I believe this is the first small step in our journey. We are currently working on a breakthrough experiment with a real quantum computer for chemistry and look forward to sharing Insilico's best practices with industry and academia,” said Alex Zhavoronkov, PhD, founder and CEO of Insilico Medicine. 

Armed with these positive results, Insilico plans to integrate the hybrid quantum GAN model into Chemistry42, its AI drug design engine, ushering in a new era of AI drug discovery. 

About Insilico Medicine

Insilico Medicine, a clinical-stage end-to-end artificial intelligence (AI)-driven drug discovery company, connects biology, chemistry, and clinical trials analysis using next-generation AI systems. The company has developed AI platforms that utilize deep generative models, reinforcement learning, transformers, and other modern machine learning techniques to discover novel targets and to design novel molecular structures with desired properties. Insilico Medicine delivers breakthrough solutions to discover and develop innovative drugs for cancer, fibrosis, immunity, central nervous system (CNS), and aging-related diseases. For more information, visit www.insilico.com