Skip to main content
Email Print Share


Computational science takes the Nobel stage

2013 Nobel Prize winners in chemistry continue to develop important high performance computing methods

Illustration of simulations of a free energy perturbation study of phosphate hydrolysis in water

Snapshots from simulations of a free energy perturbation study of phosphate hydrolysis in water.

February 11, 2014

The 2013 Nobel Prize in Chemistry was awarded to Martin Karplus, Michael Levitt and Arieh Warshel for using computer modeling to understand complex chemical processes.

Their early contributions to the areas now known as "computational structural biology" and "computational biophysics" laid the foundation for much of the current research underlying our ability to quantitatively understand the dynamics and functions of biological molecules, as well as the structure of complex materials.

More specifically, the three Nobel Laureates realized that to understand what's happening at the atomic level, the electronic motion in chemical systems must be treated using the laws of quantum mechanics (QM), but that the much heavier nuclei could be treated using Newton's classical equations of motion or molecular mechanics (MM). So Karplus, Levitt and Warshel used algorithms to develop a hybrid method, known as QM/MM, that allows scientists to accurately predict the behavior of chemical systems, a capability that is critical in biology and medicine and crucial to the development of new chemicals for industry.

Like many great ideas, the representation of molecular dynamics developed by Karplus, Levitt and Warshel was conceptually simple. There was however, one small catch. In order to perform some of these calculations, they needed the power of advanced computers.

"As one of the pioneers of using computer simulations for complex molecular systems, I learned since the late 60s to use very limited resources to capture the main physics of biological systems, without consuming enormous computer power," said Warshel.

In the mid-1980s, the National Science Foundation (NSF) began supporting advanced scientific computing, but before then, it was quite difficult for most academic researchers to get access to large, powerful computers, which were typically located at defense laboratories. When the calculations they required outpaced their personal computers, the future Nobel Laureates turned to NSF supercomputers to perform their large-scale computations.

"Computational structural biology, the field that I pioneered with Martin Karplus and Arieh Warshel, has certainly grown and matured through access to NSF-funded programs like XSEDE," Levitt said.

The eXtreme Science and Engineering Discovery Environment (XSEDE) is an environment of digital resources and help staff that allows a growing community of researchers, scholars and engineers to do their science more quickly and effectively. XSEDE users include biologists, chemists, historians, economists and other researchers that number in the thousands.

"Our 2013 Nobel Prize in Chemistry represents a huge step forward in the perception that high-performance computing is now of clear importance in a field of study previously considered as being purely experimental," Levitt continued. "The importance of XSEDE lies in its ability to work across many disciplines with a broad spectrum of users extending from novices to the most experienced users and all this at no cost of the researcher."

In recent years, Karplus, Levitt and Warshel have all used NSF-supported cyberinfrastructure to advance their research. Karplus used supercomputing resources--including the Queenbee, Big Ben and Lonestar supercomputers that were part of TeraGrid, XSEDE's predecessor--to simulate hundreds of thousands of interacting particles and shed light on the dynamics of molecular motors, like ATPases, that power the cells in our bodies.

Levitt did seminal work on the NSF-supported Cray XMP at the San Diego Supercomputer Center (SDSC) in 1986. Specifically, he and his colleagues pioneered the simulation of protein dynamics in a system of water molecules. Most modern simulations follow the protocol they defined then.

"The use of the NSF computer was essential as the computation would have been impossible on a less powerful machine," said Levitt.

Warshel was hesitant at first to apply massive supercomputing resources to his research problems. In fact, he had been very successful in exploiting simple models in a unique exploration of the action and directionality of molecular motors. However, over time, he came around to the value of high-performance systems for science.

"I started recently to accept that the use of massive computer power can now present a serious competition to my more economical direction and started to truly appreciate the need of massive computational resources," said Warshel. "Having convenient access to supercomputer resources is of a crucial importance."

He is currently using the Gordon supercomputer at the SDSC to study phosphate hydrolysis in solution and in proteins. His team's simulations led to the discovery of how the bonds connecting phosphates to other molecules in chemical compounds are broken in the presence of water.

"This reaction is arguably the most important biological reaction," Warshel said. "Using massive computer time has been absolutely essential for resolving mechanistic controversies and progressing in the field."

The molecular dynamics methods that Karplus, Warshel and Levitt pioneered have evolved into the standard approach to investigate complex chemical and biochemical processes and the behavior of materials.

"We owe it to these three Nobelists who paved the way for other scientists to now probe the details of how chemical reactions occur, how proteins fold and how complex materials assemble in nature," said Barry Schneider, a former program director for High Performance Computing at NSF, who is now at the National Institute for Standards and Technology.

XSEDE's use in classical high performance computing fields like chemistry, molecular dynamics and physics, but also economics, history, linguistics and more, continues to grow, providing the footing for future discoveries.

"Just as structural biology became accepted as a computational science, many other scientific research areas have made the same transition with the help of XSEDE," Levitt said. "It is expected that other fields of study in 'non-traditional' fields will benefit as XSEDE continues to be a resource that accelerates scientific discovery, enhancing as it does the productivity of researchers, engineers and scholars."

--  Aaron Dubrow, NSF (703) 292-4489

Martin Karplus
Michael Levitt
Arieh Warshel
John Towns
Michael Norman

Related Institutions/Organizations
Harvard University
Stanford University
University of Southern California
University of California-San Diego
University of Illinois at Urbana-Champaign

Cambridge , Massachusetts
Los Angeles , California
Stanford , California
San Diego , California

Related Awards
#0910847 Flash Gordon: A Data Intensive Computer
#0003872 Structure Function Correlation of G-Proteins
#0342276 Structure Function Correlation of G-Proteins
#0836400 Structure Function Correlation of G-Proteins
#1243719 Structure Function Correlation of G-Proteins
#9808638 Structure Function Correlation of G-Proteins
#9617544 Theoretical Methods Applied to Biomolecular Chemistry
#1053575 XSEDE: eXtreme Science and Engineering Discovery Environment
#8303385 Theoretical Studies of the Primary Event in the Visual Process
#8816740 Theoretical Methods Applied to Solution and Biomolecular Chemistry
#9119836 Theoretical Methods Applied to Solution and Biomolecular Chemistry
#8519194 Hydrophobic and Hydrophilic Forces in Membrane and Globular Proteins (Part II)
#0619926 MRI: Acquisition of a Hybrid Shared-Memory / Massively-Parallel Commodity Cluster for Cost-Effective Super-Computing at Stanford

Years Research Conducted
1986 - 2014

Total Grants