Scientists at IBM's TJ Watson Research Centre, along with a colleague from Wabash College, have published the first research results based on molecular dynamics simulations run on Blue Gene/L hardware.

Using the processing power of a half rack of a Blue Gene/L system, the group was able to increase the molecular dynamics simulation time over previous efforts by a factor of about five, while modelling a more complex biological system.

Specifically, the researchers conducted a 118-ns molecular simulation of rhodopsin embedded in a system containing cholesterol and two different lipids (each with polyunsaturated molecular chains). Previous simulations of simpler biological systems typically modeled 18-ns of an interaction. The results were published in the Journal of the American Chemical Society.

The combination of a longer simulation time and increased system complexity yields much more realistic results, which can be used to help explain the nature of the interactions observed in lab experiments, according to IBM.

Commenting on the work, IBM's Ajay Royyuru, senior manager of the Computational Biology Centre at IBM Research, noted that the scale of computing used here is unprecedented. "Problems have existed, but they've been unapproachable with existing computers," Royyuru said.

He said the processing power available in the Blue Gene/L computer allows larger, more complex, biological systems to be simulated. "This gives [researchers] more insight into the actual dynamics taking place," he said.

Royyuru will share his views on this topic at next month's Bio-IT World Conference & Expo. He will be speaking on Blue Gene/L's use in life science research as part of the IT Solutions for Drug Discovery program track, which runs from May 17-19 in Boston.