Up to 23 million people worldwide suffer from deadly Chagas and Leishmaniasis diseases, caused mainly by parasites in developing countries. Those who are lucky to be diagnosed in time survive, but all in all, treatments are ineffective with exhausting side effects. So far, medicine hasn’t come up with a good solution, and using available drugs doesn’t seem to do the trick. But there’s hope, and it doesn’t come from a Pharma company. Alphabet’s DeepMind has announced that their AI is ready to help with solutions for neglected diseases.
The tricky proteins
It was one of the most exciting announcements of 2020. Artificial intelligence cracked the mystery of how proteins fold into their typical three-dimensional structures. In fact, DeepMind’s AlphaFold opened up previously unthinkable opportunities. It helped scientists understand diseases and develop new effective drugs, even for once hopeless diseases.
“Being able to investigate the shape of proteins quickly and accurately has the potential to revolutionize life sciences,” Dr. Andriy Kryshtafovych from the University of California said for BBC back then. Without AlphaFold, scientists would have to dedicate years and exhaustive resources to decipher a shape of a single protein. Deep learning, AlphaFold’s core technology, proved to be a game-changer that does the same with exceptional results. “The 3D models of proteins that AlphaFold generates are far more accurate than any that have come before – marking significant progress on one of the core challenges in biology,” explains DeepMind’s blog.
The blog indicated that this breakthrough transcends potential health benefits because understanding protein folding enables designing proteins. As a result, we could engineer bacteria that secrete proteins to turn harmful waste such as plastic and oil biodegradable.
AI and neglected diseases
This groundbreaking innovation will now serve as hope for those affected by neglected diseases. DeepMind is teaming up with the Drugs for Neglected Diseases initiative (DNDi), a non-profit pharma trying to cure sleeping sickness, Chagas disease, and Leishmaniasis. Wired reports that DNDi with a team of researchers already have identified a molecule likely to bind to a protein on the parasite causing Chagas disease. To destroy the parasite, scientists need to understand the protein fully. While it would take years in the past, AlphaFold already armed them with data.
“This could allow us to crack Chagas disease and Leishmaniasis a lot more quickly than it looked like we were going to be able to do a couple of years ago,” said Ben Perry, a medicinal chemist and project leader at DND for Wired. “If you can quickly get these protein structures, you can design multiple drug candidates, so you have lots of shots on goal for clinical trials,“ he added.
That ties into another goal of DNDi and DeepMind, which is to help democratize the drug discovery process. AlphaFold’s data will empower scientists in developing countries to work on their own treatments, no longer depending on inaccessible resources.
Yes, skeptics warn that deep learning in medicine doesn’t automatically lead to a magical cure. There’s simply too much we don’t know yet. However, AlphaFold’s efforts sure do bring hope to areas where it was previously scarce.
Photo credit: DeepMind (a protein’s function is determined by its 3D shape)