On Sunday, February 1, 2004, at 11:20 AM, Peter Murray-Rust wrote:
[lots of stuff deleted]
Great. I'm guessing that this can be program independent and that anyone can access this nameserver whatever the platform and OS, providing they have a WWW library of equivalent (obviously this is built into java.net.*). The functionality would be:
list all the supported conventions and OpenSites. I guess these would include:
- PDB
- NCI
- NMRShiftDB
- WWMM
So you would ask for the *address* for pdb:1crn *at RCSB* and it would return:> http://www.rcsb.org/pdb/cgi/ explore.cgi?pid=179081075637979&page=0&pdbId=1CRN
(I'm hoping the pid is independent of the code)
or for NSC10 at dtp.nci.nih.gov:
http://dtp.nci.nih.gov/dtpstandard/servlet/ ChemData?queryHOLD=&searchtype=NSC&chemnameboolean=and&outputformat=htm l&searchlist=10&Submit=Submit
I am assuming these can be generated by a template and that the server would create a complete URL. It would return the URL (not its contents) to the user.
Note that mirrors have completely different formats. The main maintenance problem will be working out what the templateURLs are for all the mirrors.
I think this would be a really valuable contribution to molecular science on the web.
So I'm getting back into my proper timezone (but still longing for the 80 F temperatures) and I said I'd make some more detailed comments regarding this thread. First of all let me point out that this is especially timely considering the new NIH roadmap initiatives (See http://nihroadmap.nih.gov). The particularly applicable group is the Molecular Libraries and Imaging one. This has funding for a small molecule repository of at least 500,000 compounds, screening centers, and most relevant to this discussion, provisions for a "PubChem" effort at the National Library of Medicine, and funding for a number of Cheminfomatics centers. The details of some of this are not yet available, but it is the first time the NIH clearly recognizes that cheminformatics is not really contained in the sequence analysis that makes up a good part of bioinformatics. I imagine most of the people reading this list do not need to be convinced that this is a good thing. I also realize that some of the people reading this may be interested in competing for some of these awards and I will be sure to sent a note to this list when anything concrete is announced, especially regarding the chemoinformatics centers.
Back to the thread, which is essentially a discussion about a particular case of a very important general problem: How can a computer program get access to data from a variety of databases available online. There are two main parts to this question: 1) what are the mechanisms and syntax for making the request and 2) what kinds of requests (searches) can be made. As Peter mentions above, we have implemented a number of servlets that can be accessed via URLs with standard syntax for arguments. Typically one can search by NSC number, CAS registry number, plate number, wellid for NCI distributed plates and chemical name. The syntax for all of these searches is common to all servlets (at least it is supposed to be) and thus easy for a computer program to construct using a few simple rules. I'm very open to modifying this setup in any way the community thinks would be valuable. I'm also open to maintaining whatever kind of JMol syntax -> NCI syntax catalog that needs maintaining. We are using JMol (an old version) to display our 3D structures and for a variety of reasons we haven't kept up with JMol development. Our software development efforts (See http://dtp.nci.nih.gov/dw/dw_main.html) have reached the point where we are likely to get more involved in the current JMol efforts and we want to make our efforts be as broadly useful as possible. There's a lot more I could talk about, but I think I will stop there for now and direct anyone interested to http://dtp.nci.nih.gov/docs/dtp_search.html to see the kinds of data we have available.
DanZ
/*******************************************************/ * Daniel Zaharevitz * Chief, Information Technology Branch, DTP, NCI * [EMAIL PROTECTED] * phone: 301-496-8747 /*******************************************************/
------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Jmol-developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-developers
