Treating the Patient, Not the Illness TACC collaborates with Peter Coveney of University College London on cutting-edge HIV

Every field of study has its “Holy Grail” — a tanta- particularly vital in the treatment of HIV, where the lizing new model or discovery with the potential to high rate of viral mutation threatens to derail de- revolutionize society. With incremental advances in cades of inquiry and billions of dollars in research theory, observation, experimentation and computer and development that have led to nine FDA-ap- , researchers advance towards this goal, proved, anti-viral HIV protease drugs. pulling science and technology along with them. “Many of these infectious diseases are clever In biomedicine, the Holy Grail is patient-specific enough to develop strategies to resist the drugs medical treatment, says Peter Coveney, Professor of over time,” Coveney said. “The HIV virus is very Physical Chemistry at University College London good at doing this because of the way it copies (UCL) and Director of the Centre for Computational itself. The error rate in reproduction is so great that Science there. Instead of a generalized approach to its mutations allow it to outwit drugs.” Though illness, personalized medicine promises treatments Coveney’s research has dealt mainly with HIV, his developed just for you and sophisticated diagnostic computational approach could apply equally well tools to differentiate your genetic disposition and to the flu, hepatitis and other mutating diseases. strain of illness from your neighbor’s. Stoica, Sadiq and Coveney’s study is the most Coveney is among a handful of scientists leading compute-intensive investigation of the cause of the charge to develop patient-specific approaches saquinavir resistance in HIV protease mutants, and to medicine. His simulations of drug-resistant HIV helps to explain exactly why mutations in the HIV protease proteins, performed with the resources of protein, protease, cause medicines to lose their ef- the Texas Advanced Computing Center (TACC), fectiveness. “How can these single point mutations push the limits of biomedical research and test — some of which may be nowhere near the ac- methods for the fast and accurate treatment of tive site in the protein —how can they have such a unique illnesses. The results of Coveney’s research dramatic effect on whether the drug is inhibited or at TACC, performed with Dr. Ileana Stoica and Ka- not? There’s no way of understanding that unless shif Sadiq, were published in the prestigious Jour- you have a faithful, atomistic molecular represen- nal of the American Chemistry Society on January tation of the protein and how it interacts with the 29, 2008. drug,” Coveney said.

“Given that we know so much now about the ge- Coveney and his colleagues first developed fine-de- netic makeup of individuals and are able to collect tailed molecular representations for HIV protease data on them as individuals, not just as a ‘generic and saquinavir, the first HIV drug to be discovered patient’, the obvious challenge is to develop treat- that inhibits the replication of HIV by binding to ments that address the particular patient and the the protease molecule. They then used molecular ailments that they may have,” Coveney explained. dynamics codes that fully exploit the massively parallel power of TACC’s to deter- Though this method is relevant to all aspects of mine how well the drug binds to the “wild-type” health, the need for patient-specific medicine is HIV protease and to three drug-resistant mutant

Texas Advanced Computing Center | Feature Story For more info, contact: Faith Singer-Villalobos, Public Relations, [email protected], 512.232.5771 Page 1 of 2 strains. Stoica, Sadiq and Coveney’s results mir- haven’t done this yet, but we see a way of doing it rored the experimental data and revealed how the in the future.” virus protein interacts with the drug, including how mutations alter critical interactions to lower For Chris Hempel, TACC Associate Director for the strength of drug binding. User Services, the project was a chance to test the flexibility and scheduling capabilities of the 22nd Identifying the cause of HIV drug-resistance is largest in the world, proving just important, but Coveney’s ultimate goal is to create how powerful a tool for science it can be. “Projects a system that can recognize mutations quickly and like this have a fantastic impact,” Hempel said. administer the proper preventive drug in a suitable “Coveney’s research not only helps physicians timescale. “Can we do computations sufficiently understand the mechanics of HIV drug-resistance, rapidly and accurately and rank the different drugs but it enables the formulation of a future, computa- that might be used as treatments in terms of which tionally-based medical treatment. His work is truly will be most favored versus which will be most re- ground-breaking.” sisted by the patient? In the long run, these sorts of scenarios in biomedicine may end up being life or To date, Coveney has focused his biomedical re- death decisions, and will use computational meth- search mainly on the HIV protein, protease. How- ods,” Coveney said. “Our research is putting some ever, his future studies will model and predict the of the pegs in the ground for whether that’s viable binding energies of reverse transcriptase, another or not.” molecule involved in HIV replication that is signifi- cantly larger than protease. For these simulations, The massive scale of Coveney’s simulations neces- Coveney will likely use Ranger, the world’s largest sitated the use of high-performance computing supercomputer for open science research, which (HPC) resources like TACC’s Lonestar system to comes online in February 2008 at TACC. speed discovery. “If we want to do a 10 nanosec- ond run [the amount of time it takes a drug-protein “The huge numbers of processors that are available interaction to reach equilibrium], we have to run on Ranger make it an incredible tool. If you’ve got the simulation for 60 hours with over a million the problems, or you can think of ways of using time-steps,” Coveney said. “The sheer immensity of that scale of machine – and we certainly can – then Lonestar and its speed and power have allowed us you can have a field day there,” Coveney said. to run the simulations at TACC faster than any- where else.” At present, the realistic application of patient- specific medicine remains quite a few years away. At the Supercomputing ’07 conference in Reno last But by harnessing the power of supercomputers, November, Coveney collaborated with TACC to Coveney is blazing a path to rapid advances in test the urgent computing capabilities of Lonestar computationally-assisted diagnosis and drug pre- — its ability to turn around emergency, directed scription. simulations in real-time. Using 200 dedicated nodes over a period of six days, Coveney, with assistance “To be able to tailor medical treatment to a person from members of TACC’s HPC and Advanced and their ailments, instead of giving them some Systems groups, reproduced the conditions that a average course of treatment — we’re only go- future clinician might face when diagnosing and ing to get to that level of patient specificity if we prescribing medicine for a drug-resistant HIV pa- use computational science and high performance tient. computing, of that there can be no doubt,” Coveney said. “Computational science is not going away, it’s “The doctor would make a request and it would just going to become increasingly pervasive as time trigger a workflow running on a grid infrastruc- goes on.” ture that would, in a short period of time, bring the clinician the results,” Coveney explained. “We

Texas Advanced Computing Center | Feature Story For more info, contact: Faith Singer-Villalobos, Public Relations, [email protected], 512.232.5771 Page 2 of 2