| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: FightAIDS@Home
Thread: Improving Algorithm Efficiency of Autodock/FAAH |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Does Scripps Research Institute or Olson Labs do any further research/development to improve the efficiency of the protein-ligand free energy calculation algorithm for FAAH?
For example, there are two published papers from Folding@Home which claim improved efficiency for protein-ligand free energy calculations: Fujitsu Labs Shirts et al. used about 400 000 CPU days for the eight binding-energy calculations and obtained binding free energies with a rms error from a linear fit of about 1 kcal/mol and a maximum deviation from the fit of 2 kcal/mol. We performed similar calculations on the Fujitsu BioServer and obtained more predictive binding free energies with a rms error from a linear fit of only 0.4 kcal/mol and a maximum deviation from the fit of 0.6 kcal/mol, using very similar methodologies as the previous study, but with significantly less computational power about 50 000 CPU days. Michael Shirts thesis I also present data from various systems demonstrating that this method is in practice several times more efficient than other commonly used methods. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
http://autodock.scripps.edu/
I was about to say that AutoDock has been under continuous development since 1990, but looking at the copyright notice on the web page, I have to amend that to 1989. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Great question!
We are continually working to improve all aspects of AutoDock, including both the search efficiency and the way we score our candidate dockings. Last night, for example, I succeeded in improving the ability of the Lamarckian Genetic Algorithm to search the orientational degrees of freedom of the ligand. We are also currently working on improving the binding free energy predictions (but unfortunately, I cannot give any more details than that, until our new methods are published). These changes won't make it into the version of AutoDock running on the World Community Grid until we release a new version of the client. In the meantime, as always, thank you to everyone who donates their computer's idle cycles to FightAIDS@Home: rest assured, your donations of computer time are not going to waste. Dr Garrett M. Morris (Olson Laboratory, Molecular Graphics Lab) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
ONYA Doc.
 |
||
|
|
zombie67 [MM]
Senior Cruncher USA Joined: May 26, 2006 Post Count: 228 Status: Offline Project Badges: 
|
We are continually working to improve all aspects of AutoDock, including both the search efficiency and the way we score our candidate dockings. Last night, for example, I succeeded in improving the ability of the Lamarckian Genetic Algorithm to search the orientational degrees of freedom of the ligand. We are also currently working on improving the binding free energy predictions (but unfortunately, I cannot give any more details than that, until our new methods are published). In addition to improving the efficiency of the science, is there any work being done to improve the efficiency of the WUs, to maximize the use of the individual machine architecture? For example, the work from alexkan has made even lowly G4 macs about as productive as P4 windows machines.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Any estimate on when you will release the new agent?
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
In the meantime, as always, thank you to everyone who donates their computer's idle cycles to FightAIDS@Home: rest assured, your donations of computer time are not going to waste. Dr Garrett M. Morris (Olson Laboratory, Molecular Graphics Lab) Thank you very much for your response and words of encouragement, Dr. Morris. We always appreciate and value any updates from the project scientists. |
||
|
|
|