CSC and ChE student, lead developer of GPU Optimized Monte Carlo molecular simulation code
A chemical engineer in Iran’s prosperous petroleum industry, Mohammad Soroush Barhaghi worked on oil rigs for two years before deciding that he wanted more for himself.
He left Tehran in 2014 for Wayne State in Detroit to broaden his horizons beyond labor, beyond petroleum—and hit the ground running.
Now pursuing a master’s in computer science and a doctorate in chemical engineering, Mohammad develops complex algorithms for the sampling of increasingly complex molecular systems as the lead developer of the College of Engineering’s open-source, GPU Optimized Monte Carlo molecular simulation code (GOMC).
The software “enables scientists around the world to study relationships between atomic-level interactions and phase behavior, physicochemical properties and condensed phase structure,” says Dr. Jeffrey Potoff, the Associate Dean for Academic and Student Affairs in the College of Engineering and the chair of Mohammad’s dissertation committee. GOMC saves researchers time, money and has improved “computational efficiency compared to existing methods.”
“Before GOMC, there weren’t many tools to perform Monte Carlo simulations,” Mohammad says, “and most of those that did exist were not capable of taking advantage of multi-core and GPU architectures.”
Using GOMC, scientists can study equilibrium phenomena, nucleation, screen crystal structures to determine top candidates for alternative fuels, and even optimize drug design by simulating how drug molecules will behave in different bodily environments, Mohammad says.
“A lot of my effort has been focused on making GOMC easy for others to use,” he says.
He’s created tutorials that help others learn how the code can be used for a myriad of purposes. GOMC workshops were presented by Potoff in Mexico last May and at the annual American Institute for Chemical Engineering conference last October.
Mohammad’s proudest academic achievement so far? The development of the Molecular Exchange Monte Carlo, an advanced sampling technique that increases the efficiency of Monte Carlo simulations by at least a factor of 100, he says. “I implemented this method in GOMC and published two invited articles on it.”
Not all of Mohammad’s experiments at Wayne State have involved GOMC. At STEM Day the last two years, he’s made ice cream using liquid nitrogen. And yes, it does taste better than store bought, he says. “The trick is a surplus of heavy whipping cream.” It helps that the nitrogen freezes the milk instantly, resulting in smaller ice crystals and a creamier texture.
But it’s his work with GOMC that will prep him for his dream job of working with a pharmaceutical company that uses computational chemistry to discover new drugs.