Supercomputer Models Ribosome

Scientists at Los Alamos National Lab have successfully created an in silico molecular simulation of the ribosome, the cell's biological machine that "reads" RNA to create proteins.

Using 768 processors of the 8192 available on “Q,” the Los Alamos supercomputer, the researchers simulated 2.6 million atoms in motion for a computer model of the ribosome at work.

A paper describing the effort will appear in the Oct. 24 edition of Proceedings of the National Academy of Sciences.

This success has implications for drug discovery as well as for systems biology, a multidisciplinary and emerging scientific effort which, in part, has biologists and computer scientists cooperating to construct computer models of biological processes at the molecular level.

"The methods and implications lie at the interface between biochemistry, computer science, molecular biology, physics, structural biology and materials science," Kevin Sanbonmatsu, leader of the Los Alamos team, said in a statement. "I believe the results serve as a proof-of-principle for materials scientists, chemists and physicists performing similar simulations of artificial molecular machines in the emerging field of nano-scale information processing.

The study identified a corridor, a new region inside the ribosome, through which the transfer RNA must pass in order for the RNA decoding process to occur, potentially offering new targets for antibiotics.

While this study modeled the ribosome, researchers at Lawrence Livermore National Lab are working to join the Blue Gene supercomputer and the Purple supercomputer, formerly housed at IBM's Poughkeepsie Development Center, to model the human brain, a long-term effort conservatively estimated to cost $100 million.