Thanks for the response. Based on the pile of paper I saw being spit out of the printer, I wonder how anyone could analyze so much data. It literally was several feet tall. It seems to me you'd need another computing grid just to look at what has been spit out. At the very least, a room the size of a call center full of people reviewing the paperwork. Is that what they've got going now?

Hi,

Yes, I review the data. We discuss the progress in detail on the "Status page" and in the FightAIDS@Home Newsletters. See http://fightaidsathome.scripps.edu These Newsletters also discuss the peer-reviewed papers with FightAIDS@Home data that have been published.

Like all other labs in the scientific community, we could always use more staff. Of course we could analyze more data quicker if we had more people in the lab working on it, but funding has been tight in scientific research since ~ 2000. In addition, there are not that many people in the world who have been trained to do what we do (and are willing to do it for the salaries offered in academia). We work at a non-profit research institute, not a large pharmaceutical company. If you think we should have a larger lab, then contact your representatives in Congress and ask them to increase the NIH budget for disease-related research. You could also encourage people to donate to The Scripps Research Institute, which will help advance the research against many different diseases.

None of the FightAIDS@Home data is wasted. The data are stored electronically. Unlike a carton of milk or a stack of Petri dishes, the FightAIDS@Home data does not have a "shelf life." It won't go bad. I analyze the data one experiment at a time. As soon as the results arrive from WCG, we run a "quality control" process to check the integrity of all of the results, and then I immediately process all of the results (that is, I cluster the results to determine how many of the independent docking runs placed the ligand in the same region with the same binding mode, I measure the types of interactions the ligand forms with the target, and I collect statistics on the energetics and reproducibility of these different binding modes). You are correct that we do not visually inspect the binding mode in each and every result. No virtual screeners do. We focus on the "cream of the crop." Instead of spending a lot of time looking at the hundreds of thousands of ligands that do not dock well to the target, I visually inspect the binding modes for the few hundred compounds/experiment that displayed the best energies and the best statistics. It's like looking for a needle in a field full of haystacks. It takes time to find the best results, and it takes more time for these computational predictions to be evaluated in test tubes.

All FightAIDS@Home data is public domain, as well. If other HIV researchers want to analyze our FAAH data, they just have to ask for a copy. There are no secrets here.

It took a while to get the logistical issues figured out and to optimize the workflow. I've been part of the FightAIDS@Home team since August of 2007. Since I started leading the day-to-day operation of the project, we have been analyzing more and more of the data and discussing our progress in more detail. But you are correct that the amount of data generated on FightAIDS@Home is massive. The necessity to analyze even more of the data even quicker has become the mother of invention: Stefano Forli (a member of the FAAH team) and Ruth Huey (another member of Prof. Art Olson's lab at TSRI) have created a new tool called "Fox" for analyzing massive virtual screens in a much quicker, more thorough, and easier way. After Stefano and Ruth hone this tool a little more, I will start using it to analyze more of the FightAIDS@Home results much more quickly. This new tool will not only accelerate our analysis of FAAH data, it will also help increase the efficiency of many other labs in the structure-based drug discovery and design community.

Sincerely,
Alex L. Perryman, Ph.D.

Hi,

I'm involved in this project for years, now. I usually don't post that much on the forum, since Alex already does a great job with it. I work side by side with him to dig the data looking for interesting results and in developing the software we use daily to analyze the docking results.

Since the project began, we've had to manage a huge amount of information, and this challenging task was a great opportunity to develop new tools for handling the FAAH data.
The results of these development efforts have been included in the AutoDock tools and are available to everyone at our web site http://mgltools.scripps.edu/

At the time you refer, there were already automated procedures to pre-process the data without requiring user intervention. In this way we could focus our attention on filtered, enriched results obtained by refinement of the raw data provided by the volunteers. We are now adding a graphical user interface to these tools, including even more functionalities directly following up from our experience with the FAAH project. Again, all these tools will be released for all the research community to benefit from it.

With the experience we gained, efficiency in analyzing the data increased dramatically requiring even less effort on our side. Moreover, new FAAH experiments are designed on the basis of findings from previous ones, therefore data analysis is a crucial point for making the next steps.

We always have a lot of work to do (your fault: we're the most popular WCG project, thank you guys!) and we never have time to wait for new results to come, but this doesn't mean we are wasting your and our time.

Our plan is to keep moving, trying to push the boundaries of Science a little bit forward to help to find a cure for this disease.

With your help.

Stefano


p.s. : ...paper, uh?

Alex -


I was part of the team back in 2004. Let me reiterate something that may have been misunderstood. In my original post I commented that my information was from nearly 6 years ago.


Actually this is incorrect. I didn't get anything confused. The team was only 3 people at the time. Dr Olson, a male who's name I forget with a british accent, and one other. That male was the head programmer for the project and handled the data. It is he who physically pointed me to the pile of paper in the back room which was nearly 4 feet high, sitting in the dark, with the lights off, and told me: "There is all the data - and we have nobody to analyze it". It was the topic of a 15 minute conversation and you can believe I was confused, even back then, why it was being kicked out in printed format. Finding a way to fund the analyzing of the data was what they were seeking to resolve. They had decided to bring in an "ideas guy" who's job it was to come up with ways to expand the project and attract investors. It was at that time that I bowed out of the process because I felt saddened the data wasn't even being looked at.

Keep in mind this is at least 3-4 years before you arrived. So you may be getting the timelines confused.


No, it wasn't Dr. Olson's desk. Unless Dr. Olson's desk was on the floor of the back room in the dark, with a pile nearly 4 feet high being spit out of a printer in real time. And the british guy lied to me :)

The bottom line is that the data is being processed now.

I am happy to hear that.

HLT