The worldwide threat from AIDS is well-known, if not always well-understood. The disease does not get as much media attention as it once did, but it remains a serious and growing health threat in most parts of the world. According to the most recent data from the World Health Organization
, there are currently about 35 million people worldwide living with HIV infections; two million people are newly infected and over a million die from the disease each year. However, simply being infected by the virus is different from having the disease. HIV (the human immunodeficiency virus) can eventually cause AIDS (acquired immune deficiency syndrome), a condition in which the immune system is weakened so much that the patient becomes ill and dies from some other causes that the immune system would ordinarily be able to control.
The treatment landscape
There have been tremendous advances in treatment of HIV since the virus was first identified in 1983, and with modern antiretroviral (ARV) drugs, people infected with HIV may be able to delay the onset of AIDS for many years. However, there is no cure, and the most effective treatments are extremely expensive and do not work for everyone. Less than half of the people who could benefit from antiretroviral therapy currently receive it. For all of these reasons, research into potential treatments for HIV has been a topic of intense interest for decades, and continues to be today.
HIV remains a difficult virus to stop because when it replicates, its genetic material (RNA) is often copied imperfectly, and as a result the virus is continually mutating and changing. The mutated viruses that can survive in the presence of known HIV drugs can thus become "drug resistant." One promising area of research is to identify and target areas of the virus's proteins, encoded by its RNA, that remain relatively unchanged over time - the hope is that once a treatment is found, it will remain effective for a long time. The challenge is to find chemicals that will bind to these specific target areas and inhibit the virus without disrupting other systems in the patient's body.
New tools mean new hope
Computer simulations offer an exciting and increasingly effective tool to help accelerate this search, because researchers are now able to scan entire libraries of chemicals and predict which ones might form the basis for effective drugs. This pre-screening can reveal where best to focus development efforts, dramatically reducing the amount of lab testing required and (in theory) making new drugs easier and faster to produce. However, there is always a balance to be struck between simulation accuracy and simulation difficulty - large-scale simulation efforts might not always produce the most accurate results and may produce 'false positives' (where they inaccurately mark a chemical compound as promising).
Thus, while researchers face huge challenges, there is also huge potential. Computer-based simulations can identify promising chemicals for further study and drug development, but sometimes the "false positives" end up wasting time and money. Even with all these new tools and techniques, the process of drug discovery remains an extremely long and expensive one.
The Proposed Solution
A decade in the making
The first phase
of the FightAIDS@Home project was an important part of the potential solution. Over the past 10 years, volunteers contributing to FightAIDS@Home have screened millions of chemical compounds, looking for those that might bind to specific sites on HIV and help disrupt the virus's lifecycle. The project has had some remarkable successes
, and has expanded over time to screen new compounds, investigate new potential binding sites on the virus (on protease, integrase and reverse transcriptase), tap into new sources of computing power (Android mobile devices), and use new computational screening programs (AutoDock Vina) to improve the performance and accuracy of the simulations.
More accurate simulations
Phase 2 introduces a completely new computational tool into the FightAIDS@Home effort: the Binding Energy Distribution Analysis Method (BEDAM), implemented with the Academic IMPACT molecular mechanics and molecular dynamics engine software. The BEDAM approach models the reorganizational energy of the interaction of the docked complex; it is a proven technique that has been shown to work on test data where the true results are known, but one that has never before been implemented on such a large scale. Put another way, Phase 2 uses a different simulation method to double-check and further refine the AutoDock and AutoDock Vina screening results that were generated in Phase 1. The main goal of Phase 2 is to help identify false positives - chemicals that the initial screen suggested would bind to the target site, but which actually won't work when tried in the lab. BEDAM is well-suited to this kind of work because although it has proven to be more accurate than AutoDock and AutoDock Vina, it needs an already docked complex to evaluate, and uses much more computer time to evaluate each docked complex. Therefore, it makes sense to apply the technique to only the top results from the Phase 1 virtual screen.
The overall goal of FightAIDS@Home is to find new leads for HIV therapeutics. Our work is the first step in a long pipeline that may eventually lead to a new drug or therapeutic strategy. However, Phase 2 is focused on two goals in particular:
1) Reduce wasted time and money in the lab testing stage of research
The main goal is to use BEDAM simulations to verify the most promising compounds that were identified in Phase 1 screening, and eliminate false positives. By adding this additional computational step, there is a much better chance that the eventual top-rated compounds will actually perform well in lab tests.
This is important because of the cost of laboratory testing. After the two-step virtual screening and BEDAM/Academic IMPACT analysis to identify "hits" (compounds that show promise for further development), the best compounds identified would then need to be either purchased or synthesized to be tested experimentally. This laboratory testing is an expensive and lengthy process, but by using BEDAM we hope to limit the number of false positives and narrow down the list of promising results to save time and money as the compounds move into the drug development pipeline.
2) Expand the arsenal of virtual screening techniques for research on other diseases
The methodology that we deploy on HIV drug discovery can be and is applied to other diseases. The docking tools we deployed and used in Phase 1 (AutoDock and AutoDock Vina) have both been used extensively by scientists worldwide to screen chemical compounds as potential treatments for cancer, tropical diseases and other health issues. Over 5000 scientific papers have been published using these tools. Our use of BEDAM to analyze pre-screened compounds at this scale is unprecedented. If this approach yields the results we anticipate, these research techniques will benefit other areas of research, including other World Community Grid projects.