Artificial Intelligence for New Drug Design

We talked to Yann Gaston-Mathé of Iktos on how to develop an innovative and user-friendly technology platform for deep-learning based de novo drug design and this is what he had to say.

First of all, how are you and your family doing in these COVID-19 times? 

Yann Gaston-Mathé: Thank you for the kind question. Fortunately, we are all doing well, including my parents, who are more at risk. Let us keep our fingers crossed!

Tell us about you, your career, how you founded Iktos.

Yann Gaston-Mathé: My initial training was in mathematics and biology (MS) at Ecole Polytechnique and AgroParisTech, before turning to pharma R&D at Servier. I have acquired 20+ years of experience in different life-science related sectors in Pharma (Servier, Ipsen), molecular diagnostics (IntegraGen), strategy consulting (Capgemini Consulting, BearingPoint, Cepton), and data science (Quinten). My eclectic background across science, IT, management, and business has given me a broad knowledge of the pharma industry’s business and scientific challenges and an extended professional network. My motivation as an entrepreneur was strong, and I created two biotech start-ups to develop new oncology drugs. In 2015, I met the two other co-founders of Iktos, Quentin Perron, a medicinal chemist, and Nicolas Do Huu, an AI specialist. They had started developing a new technology, based on deep learning, to automatically design promising molecules to speed up new drug discovery programs. I decided to partner with them, and we founded Iktos in Sept 2016. Since then, the company has become one of the world leaders in AI for chemical research and new drug design.

How does Iktos innovate? 

Yann Gaston-Mathé: Iktos was one of the pioneers in applying AI technologies called “deep generative models.” The initial use in natural language processing and image processing for the automatic design of novel chemical compounds with high probability to meet all the success criteria of a small molecule discovery program. This new technology can trigger significant productivity gains in the upstream pharmaceutical R&D pipeline by saving costs and time and increasing the research’s probability of success. Since our initial works, we have brought continuous improvements to the technology to improve the output quality and enable it to address more use cases in small molecule drug discovery. We have also developed an innovative technology for retrosynthesis of chemicals, which complements our automated drug design technology nicely.

How the coronavirus pandemic affects your business, and how are you coping?

Yann Gaston-Mathé: Most of our business requires attendance at scientific conferences and trade shows, which, of course, have become impossible with the pandemic. Fortunately, many of them have become virtual, allowing us to stay in touch with our partners and clients. We can also rely on our IT infrastructure and video conference solutions to move ahead in our projects and develop new solutions for our users. As a company involved in new drug discovery, we are highly sensitive to the impact of COVID-19 on our society. It pushed us to collaborate with SRI International to combine artificial intelligence and a novel automated discovery platform to accelerate the development of new anti-viral therapies against COVID-19.

Did you have to make difficult choices, and what are the lessons learned?

Yann Gaston-Mathé: We have been lucky in that regard. The coronavirus pandemic’s impact has not significantly impacted the development of the company and has not forced us to revise our strategy.

How do you deal with stress and anxiety?

Yann Gaston-Mathé; Being co-founder and CEO of a start-up company are indeed very stressful. To cope with that, I try to relax with my family (I have four kids!) during the weekends. When things get more complicated, I try to remember that I am still fortunate to be engaged in such an exciting adventure. Even if we failed, it would always have been worth it!

Who are your competitors? And how do you plan to stay in the game?

Yann Gaston-Mathé: We are involved in a very competitive field, with competitors worldwide, such as Exscentia, Insilico Medicine, Atomwise or XtalPi, and some indirect competition by prestigious academic groups such as MIT or research in AI performed by the big pharma companies. It is tough, but it is stimulating! We have a multi-pronged approach to stay in the game: 1) we focus on our strengths and differentiators. For example, we are the only company that masters generative design and retrosynthesis, enabling us to have outputs of very high quality. We keep striving to improve the technology, expand its use cases, and make it industry-ready – this is through a considerable and constant R&D effort, fueled by our exposure to real-world projects thanks to our collaborations with pharma. We want to be known as a company with whom it is easy to do business with and to work with, and who wants to make its technology widely available to the community through the development and sales of software that will be easy to use by end-users without expertise in AI or programming.  

Your final thoughts?

Yann Gaston-Mathé: Covid-19 vaccines research & development has been incredibly efficient to deliver vaccines on the market, and hopefully, we can hope the situation will normalize in 2021. It is a fantastic achievement of pharmaceutical R&D. It gives hope!  

Your website? 

https://iktos.ai/