With AlphaFold, scientists design more effective drugs than ever before
Karen Akinsanya is President of R&D, Therapeutics, at Schrödinger in Novel York. She shares her AlphaFold story.
I have always been fascinated by the idea that you can go from the bench to the patient’s bedside.
I worked in academia and drug discovery and development. That means I not only studied proteins and genes and understood how to create a therapeutic molecule against a disease-causing target, but I was also at the patient’s bedside when they were receiving that modern drug.
But the real question is, how can we improve the way we do it? People are still dying from cancer and heart disease every day, waiting for us to find solutions.
I always say that Mother Nature is thrifty. When you come across a modern drug target, you often find other potential targets that are like brothers, sisters, cousins. Each target is a protein on the surface of a cell that the drug binds to, called a receptor. The challenge for people working in drug discovery is to find a drug or molecule that binds one member of that family—the target—and inhibits that member of the family but doesn’t inhibit the rest of the family. That’s partly where AlphaFold has proven so great for us.
In some cases, AlphaFold—combined with our own physics-based software that simulates interactions between atoms—allows us to begin to simulate not only what individual family members do, but also how different family groups behave.
One of the most vital things in drug discovery is the ability to design a selective drug that affects only one specific target.
This is particularly vital when designing drugs for our endocrine system, which is regulated by the pituitary gland and a number of hormones and messengers circulating in the body.
Our recent work has looked at ways we could design molecules that selectively replicate the actions of natural transmitters, particularly those that affect how our brains function. In this case, we’re focusing on neurotransmitters in a range of neurological diseases, including conditions like schizophrenia and bipolar disorder.
What we found to be electrifying is that using AlphaFold’s predictive structures together with our structure refinement and drug design software potentially allows us to design a molecule that targets the family member of interest—in this case, the signaling partner in the brain—with extraordinary accuracy. That’s something that people have been trying to do for years and that has a whole host of potential treatment outcomes.
But at the end of the day, it’s only when this medicine reaches a real person that you’ll see what the culmination of all this research is. Will it really make a difference?
Since James Watson and Francis Crick discovered the double-helix structure of DNA, we have sought to better understand the fundamental elements of how and why our bodies work, and how and why they sometimes fail. As a human race, we ask ourselves: what is the code of life? When it comes to drug discovery, AlphaFold is an additional piece of the puzzle. Each of the elements needed to design drugs is incredibly complicated. But putting a few together can accelerate drug discovery in ways we have never seen before.