Using AI to leverage existing data to get new medicines faster

L’NGO Medicines for Neglected Diseases initiative., which seeks cures for neglected diseases, in April launched a partnership with BenevolentAI, a UK company working to develop new molecules using AI. BenevolentAI is not on the first try. She has notably shed light on the role that a molecule, baricitinib, developed by the Eli Lilly lab for another disease, can play in treating Covid-19 patients during the pandemic.

One might think that AI is now on everyone’s lips. But in pharmaceuticals, the change is more than just cosmetic. In early 2020, Exscientia, a Scottish start-up, developed a first molecule with the Japanese pharmaceutical company Sumitomo Dainippon.built” by AI, entry into clinical trial.

It’s not futuristic: artificial intelligence is a methodical approach to data processing that can be used at multiple stages of the drug industry development process.“, says Dr. Thomas Borel, director of scientific affairs at the Association of Pharmaceutical Companies (Leem).

A visit to the Parisian premises of the French start-up Iktos, founded in 2016, allows us to understand the change of times. There are no microscopes or biological apparatus here, no lab workers in white coats. But computer screens galore, traversing masses of health data at speeds no human brain could match.

The idea is to use existing data to get new interesting molecules more quickly“, explains Yann Gaston-Mathé, head of the start-up he co-founded in 2016.

His team used a global database with data on 100 million molecules. From this, “we trained a model that automatically generates new molecules and identifies those that are active at biological targets of interest,” says Yann Gaston-Mathé.

Iktos has even set up a platform for researching molecules using artificial intelligence, which it makes available on a subscription basis to pharmaceutical companies.

We use so-called generative artificial intelligence

Aqemia, a start-up founded in 2019 by the École Nationale Supérieure-PSL, is developing a drug discovery platform that leverages quantum-inspired statistical physics.

We use so-called generative artificial intelligence“, emphasizes the founder, the researcher Maximilien Levesque. “We invent molecules that stick to a specific biological target that is responsible for a disease. Artificial intelligence is powered by physics: we only need to know the physical nature of the molecule and the target to calculate their affinity“, he describes.

If start-ups are at the forefront, the large laboratories are increasingly dealing with the topic and paying the price. American giant Bristol-Myers Squibb last year signed an agreement with Exscientia for which it could pay more than $1 billion. Gafam are also involved: in 2019, the Swiss laboratory Novartis and the giant Microsoft announced their collaboration on the subject.

Is this the end of the chemist in his laboratory? There are major challenges, including access to actionable data. Without forgetting the need to find future data specialists, experts in both artificial intelligence and pharmacological topics.

There is also an important regulatory aspect, Judge Thomas Borel of Leem. “For example, AI is used to create a virtual patient arm during a clinical study. But for this drug to be accepted, regulatory systems need to recognize the value of the algorithm.“, he says.

Medicines have been developed with the help of computers for years“, for his part, comments Yann Gaston-Mathé, who says he wants to bring”additional tools for chemists without wanting to replace humans with machines“.

With AFP

#leverage #existing #data #medicines #faster

Leave a Comment

Your email address will not be published.