Novartis Enables AI Scientists to Accelerate Discovery and Development of Innovative Medicines

By Kelvin Mmbihi Lumasia

                       

                   Norvatis scientist inspect medicine.

Here's a one-of-a-kind culinary story. Indeed, chefs are chemists, ingredients are molecules, and the main course is a new medicine designed to beat disease.
At least that's the description of Luca Finelli's collation to explain in layman's terms how scientists at Novartis are researching innovative drugs based on artificial intelligence (AI), as part of a collaboration with Microsoft to deliver medicines to patients more quickly.
But this recipe depends on scientists' ability to predict what mixture of molecules can be turned into drugs, a tedious process that traditionally takes decades and can cost billions.
“Creating a drug formulation is a bit like cooking,” says Finelli, vice president and head of ideas, strategy and design at Novartis, a multinational pharmaceutical company based in Basel, Switzerland.
“As a general rule, the formulation researcher has to decide, 'I'm going to take this amount of this ingredient A and a certain amount of this ingredient B”. So they try different combinations, ”adds Finelli.
Each molecular combination must then be tested for its effectiveness, stability, safety and more. These experiments can take years to complete. And the most promising drug candidates fail somewhere along this long journey.
But by harnessing the power of artificial intelligence in collaboration with Microsoft, researchers at Novartis could reduce this process to weeks, if not days.
Comment Tools using artificial intelligence can quickly sift through data archives and the results of decades of laboratory experiments and suggest molecules with desired characteristics optimized for current medical practice. These drugs could then be rushed for further testing and, if found to be safe and effective, potentially developed and produced as a cure for the disease. This AI-powered process could eliminate years of testing with less than ideal molecules. In fact, this functionality has already been "integrated into the decision support system in front of our pharmaceutical chemists," says Shahram Ebadollahi, head of data and AI at Novartis.

The potential human impacts are vast, says Ebadollahi.
“If you look at all aspects of the pipeline - from early drug discovery and development, to clinical trials, and then to large-scale drug production - in 2020 alone, our drugs have reached nearly 800 million patients in the world, ”says Ebadollahi. To accomplish this feat, scientists at Novartis are creating molecules that have never been manufactured, and these molecules will help develop new drugs to fight diseases for which there is no cure, says Karin Briner, head of global discovery chemistry at the Novartis Biomedical Research Institutes.

A Novartis scientist inspects a drug vial. The foundation for this work is the 2019 strategic partnership between Novartis and Microsoft to "reinvent medicine" by founding the Novartis AI Innovation Lab. The goal of this alliance is to help accelerate drug discovery for patients around the world. cutting-edge technological platforms.
“Microsoft brings two things,” says Chris Bishop, laboratory director for Microsoft Research Europe.
“We bring our machine learning experience and large-scale IT. These do not exist in the pharmaceutical world. And Microsoft can't hire him (independently).We are not a pharmaceutical company, so the partnership is absolutely crucial, ”explains Bishop.” This is how the disturbance will develop. This collaboration is at the heart of everything. "
Machine learning is a fundamental building block of artificial intelligence, which enables computers to use algorithms to find patterns and trends in huge data sets.

At Novartis, researchers can apply the artificial

intelligence to examine a range of lab data from thousands of past drug development experiments: findings buried in PDF files, Excel tables, and written descriptions of the chemical properties of previously explored molecules. “Normally, they do. do it manually, reading all of these documents to find out what is relevant to the issue they have in mind,”says Finelli.
"Here, AI can actually help you do that with a few clicks and bring relevant information back to the user for later use, informing them of how to design future experiences to find new ways to create a formulation for a new drug ", adds Finelli. Novartis researchers also rely on Microsoft Azure in their work.

Thanks to artificial intelligence, researchers can now simulate thousands of experiments simultaneously.

Norvatis Laboratory researchers at work

Ultimately, scientists at Novartis aim to use computer models to help predict promising molecular structures or reveal which experiments might be most useful for testing, maintaining quality, and shortening a testing process that can now take time( years).
"Now you can do 10,000 experiments at the same time, get the result