Molecular Matchmaking for Drug Discovery

By Aaron Dubrow in Research, Special on May 30, 2012 at 9:30 am | No Comments

Some of the greatest advances in pharmaceutical history are thanks to educated guesswork or blind luck. Alexander Fleming discovered penicillin when he noticed that some accidental mold also killed bacteria. Aspartame—the world’s most popular sugar-free sweetener—came about because a chemist made the questionable choice to lick his finger in the lab.

Now there’s a better way. UT computer scientists are using the supercomputers at the Texas Advanced Computing Center to revolutionize the way new drugs are discovered. Their advanced digital models simulate how medicines will work at the molecular level—and that speeds up the long and complex process of getting a drug from lab to patient.

“Computers are a good way to accelerate the process of drug design,” says UT computer scientist and computational visualization superstar Chandrajit Bajaj. “It takes 10 years to proof out a drug, and a billion dollars or more. Hence computational drug discovery is not only timesaving, but economics tells you this is the way we should be going.”

The first step toward discovering a breakthrough new drug is to analyze the source of the illness (a virus, bacteria, or genetic mutation). Researchers shoot powerful x-rays through a sample, resulting in nanoscale pictures of molecules in near-natural conditions. But these pictures are a speckled mess that has to be cleaned up to be useful.

That’s where the supercomputers come in. Digitally, they combine 100,000 or more nanoscale pictures, resulting in a 3-D model that shows a molecule in detail. Those details matter because drugs work when molecules bind together like puzzle pieces—so researchers must study the exact shape of each. UT scientists use the TACC supercomputers to build more specific, detailed models than ever before. The new models allow them to test out every factor at play in drug binding.

“If you don’t get all the factors into simulation, you get the wrong answer and your predictions suffer,” says Bajaj. “And if your predictions suffer, you haven’t done anything to accelerate the solution. You’re just loading yourself down.”

Once scholars figure out a molecule’s structure, they test potential drug compounds to see if any of them fit a binding site—like trying puzzle pieces in search of one that fits.

Bajaj has studied the HIV virus, and his goal is to find a molecule that can bind to a specific location on the virus. The potential drug would induce the virus to harmlessly spill its contents outside the cell. New techniques like this hold hope for the 34 million HIV-positive people worldwide.

Computational drug discovery is a hot topic in academic research centers and industry alike. As chair of a study section of the National Institutes of Health, Bajaj often speaks with individuals from the pharmaceutical industry about changes in the field.

“More and more, they’re moving into the computational drug screening arena, and more and more it’s teams of people working together,” Bajaj said. “The biophysicist, the biochemist and the synthetic chemist are sitting together with the computational expert, and they say it’s giving them clues as to what they should be doing next.”

Bajaj and his team have improved the resolution and accuracy of the drug models tremendously. They’ve also sped up the search by an order of magnitude.

“What used to take months is now taking a few days,” Bajaj says.

This story first appeared on the Texas Advanced Computing Center website.

Top photo: A model of a DNA-binding protein. By Leighton Pritchard, Flickr Creative Commons.

Inset photo: The human ribosome. Illustration by the Center for Computational Visualization.

Tags: computer science, supercomputing, tacc, texas advanced computing center