Dayle,

Those percentages are what the scientists already have in their possession.
For example, while we have already crunched batch Z and are in batch A2 the scientists may still have only received up to batch X.

Actually, if Falconet would read back recent tech posts, he'd find that the traffic back and forth between Scripps/Olsen and WCG is pretty constant... an X number of batches is fetched daily, and results are send back. The status experiment page update is manual, per Dr Santiago he's trying to do this bi-weekly IIRC.

As for what's up with the weighting? [Speculation switch on]: 64 bit devices, particularly Linux are superior at processing VINA versus AutoDock [me and many have been saying this from observation for a long time... it goes for all sciences on AD VINA in fact]. There being few experiments currently for the AD engine versus many for the AD VINA, by controlling the share it works to optimize efficiency SANS having the project selection option to choose one of the 2 engines to contribute to FA@H. [In fact my W7-64 Octo is much better at the AD jobs, so it's very poorly utilized at this time... all the more reason for these 2 special hire web-interface designers to pull a fast-forward one on getting us this more granulated sub-selection option, or finally implement an automated task-type assignment that is working off efficiency statistics... with opt-in of course, the Nespresso add says here "what else?"].

(Sadly 2 of the BOINC Workshop13 vids I tried to watch, the Dr.A and the K&K presentations in avi format, are both soundless on 2 machines, so was not able to listen in what the latest 'development' directions are in the WCG box... one way to find out what's on, my English lip reading not all that advanced ;)

edit: spellcheck.

I would like to ask what is the status now? It seem like keep on getting WU that is beyond what stated in the faah status website.

Hello!

I haven't forgotten about you (in terms of updates, etc.): I will try to address in this post some questions that I have read in this thread.

I will look into particular issues about distribution of work among AutoDock and Vina jobs, but the simple fact is that Vina is faster. As some of you may know, 1 AutoDock WU is a fraction of an AutoDock job, while 1 Vina WU contains multiple Vina jobs. Also, several WCG projects have completed earlier this year, so a good percentage of donors have been contributing to FAAH.

We have been feverishly keeping up with the queue, and to date, I don't think there has been much of wait in terms of streaming AutoDock/Vina jobs. (Guiness World Record?) To give (yet more of) an idea of Vina's speed, when I first arrived (May 2013), Dr. Perryman had submitted Vina jobs numbering in the 45000's, and I have recently prepared Vina jobs through the 373000's.

A newsletter is being written and will have more details, but we have been working with WCG to try out the flexible docking feature for Vina (Experiments 97-102). Those WUs have not been submitted until testing is complete, so the batches for Experiments 103-108 may have already been sent to donors. Further, some recently submitted AutoDock jobs (Experiments 72-78, AutoDock) should give us an idea of how to increase accuracy in our dockings once all of that data has returned, and we have had a chance to analyze.

Additionally, Experiments 103-108 will aid in developing protocols to choose the best receptor structures in which to dock. You'll notice that in these experiments, there are several duplications of a PDB ID (e.g., 3AV9) along with 2 versions of each site (FBP, LEDGF, and Y3). WCG and its donors has provided an excellent opportunity to computationally investigate these concepts at a massive level WHILE working towards our original goals at the HIVE .

Ok, we have to save some details for the Newsletter. As always, thank you for your contributions.

Sincerely,

Dr. Santiago

When I download, let say, 150 WU's, only 5 - 10 are AD...