Zusammen

Free Web-based databases for chemistry are just the beginning; with a good underlying data set, many interesting and useful services can be created. A case in point is DOCK Blaster, today's stop on our continuing tour of free chemistry databases and Web services. DOCK Blaster is a unique service that attempts to make small molecule-protein docking simple enough for nonexperts to use. From the recent J. Med. Chem. Article:

Molecular docking is the most practical approach to leverage protein structure for ligand discovery, but the technique retains important liabilities that make it challenging to deploy on a large scale. We have therefore created an expert system, DOCK Blaster, to investigate the feasibility of full automation. The method requires a PDB code, sometimes with a ligand structure, and from that alone can launch a full screen of large libraries. A critical feature is self-assessment, which estimates the anticipated reliability of the automated screening results using pose fidelity and enrichment. ...

DOCK Blaster exposes a remarkably simple interface to a complex system of underlying software. The process consists of as few as four simple steps:

1. Identify a protein crystal structure. A simple way to do this is to search the Protein Data Bank (PDB). Once found, you can use the PDB code to begin your docking experiment.
2. Enter the PDB code into the online form. Don't forget to enter your e-mail address, as this is how the system will communicate the progress of your job with you.
3. Submit the form by clicking the "DOCK!" button.
4. A subsystem called "Scrutinizer" will look at the job you submitted and give you feedback as to its suitability. If everything is okay, you'll see the other "DOCK!" button. Pressing it begins your job.

At this point, most of your interaction with the system will be through e-mail. You'll get updates at the various stages of your job including target prep, calibration, and docking. Be prepared to wait; a run with the crystal structure for Fatty Acid Amide Hydrolase (PDB code 1MT5) took well over an hour to complete.

When the entire analysis is done, you'll be e-mailed a link from which you can view the results of your docking request. These results can either be viewed online, or exported in both PDB and SDF format for use with other applications. Here are my results. Results from other users are also available here.

When viewing results from the DOCK Blaster webpage, clicking on a 2-D structure takes you to a page in which a 3-D representation of both the small molecule and its immediate protein environment can be viewed interactively. There are links into the ZINC database compound records on which dock blaster is based. Another viewing option, provided you've set up your system to allow it, is to view each docking pose in Chimera or PyMol.

In addition to the quick experiment outlined here, it's also possible to specify a docking job in more detail. For example, you can indicate a binding site and supply your own list of actives and in actives.

DOCK Blaster is overall a very well-designed system that almost any chemist or biologist can start using immediately. But there are some areas that could use attention. One of the most important has to do with the long-running nature of docking jobs. Given the length of time required for a single analysis, it would be helpful to get intermediate feedback along the way. One approach might be for DOCK Blaster to return a partial list of results (or some meaningful metric) with the option to cancel the job or refine it.

DOCK Blaster is the first publicly-available service I'm aware of in chemistry that exposes a simple Web interface to a very computationally intensive task. The results it returns may not be especially useful at this point, depending on how and what the service is used for. DOCK Blaster's creators themselves note some limitations (and some remarkably good results). But in terms of using the Web to remove barriers now in place when performing certain kinds of analyses in chemistry, DOCK Blaster is on the leading edge of what will hopefully become a new trend.

Kudos

• Removing barriers to using powerful software.
• Simple interface and good use of e-mail for long-running tasks.
• Live help online!

Ideas for Improvement

• Offer a preview of results before each stage is complete.
• Expose publicly-viewable jobs as an RSS feed.
• Give an estimate of how long a job will take at the time that it's submitted.