Zhang Server for Molecular Replacement
I have mentioned the Zhang server before and so thought I would share 2 ways this server can be helpful to crystallographers.
The Zhang Server (I-TASSER server) uses the amino acid sequence of a protein to predict its structure. The server is simple to use– enter in your desired amino acid sequence and email.
Note: You need to have an academic email address to use the server freely. (more information about the server).
Good:
1) Great resource: the server allows users to submit their sequences to be analyzed
2) Free to all academic users
Bad:
1) Turn around time: it took about two weeks before my predictions were complete (this will vary greatly depending on how many proteins are in the queue upon submission)
The two ways this server could be helpful:
First
I was able to find another protein that may have a similar fold to the one I am working on. Were it not for this server, this protein would have gone unnoticed. The Zhang server searches through the PDB looking for homology that you may miss.
Second:
In silico molecular replacement? Could the coordinates from the Zhang server then be used for molecular replacement? This idea is being explored by MR-CAFASP. I have not heard of many crystallographers testing whether in silico molecular replacement would work.
Has a novel protein been published using in silico predictions? In the future, I will play around with the coordinate file from the Zhang server to determine whether or not the results are robust enough to yield reasonable crystallographic statistics.
WTF
April 23rd, 2009 at 7:32 PM #
Why use a server that doesn’t allow you to download the resulting pdb? This makes the server almost useless. I wasn’t able to find any NIH funding in his name for this research or I’ld consider writing them.
admin
April 23rd, 2009 at 8:22 PM #
Strange, I am now able to download the pdb files without a problem. The website was just redone and maybe they fixed the problem (or somehow I couldn’t figure out how to click the download button).
update: able to download the files fine windows, but not linux… anyone else having the same issue?
Prp
April 26th, 2009 at 7:55 PM #
I am a regular user of I-TASSER server and I never had this problem.
To download the models you need to click on the link that is below the images…below the C-score. I guess you are clicking on the images ….
Anyways if your case is special, you can mail to zhanglab@ku.edu, they are friendly and are willing to fix the problem (if any).
Oliver Clarke
May 4th, 2009 at 7:48 PM #
Regarding in silico molecular replacement – the work of David Baker and colleagues in this area is quite exciting.
They’ve successfully used ab initio models to phase proteins with no detectable remote homologs (though to be fair, it does require exorbitant computing power, and it only works for small proteins, and it doesn’t work very often). Very impressive nonetheless, and hopefully a sign of things to come.
See doi:10.1038/nature06249 (initial Nature paper) and doi:10.1107/S0907444908033192 (followup in Acta D)
Sean
May 6th, 2009 at 12:38 PM #
Very exciting. I did come across David Baker’s work when preparing this post. He is definitely one of the best when it comes to using ab initio models for phasing.
Thanks for bringing up the terminology used in this field. I wonder if I coined the term ‘in silico molecular replacement’, hehe.
Sean
May 17th, 2009 at 9:10 PM #
@Prp: I am still having issues viewing the PDB using Linux. I have also been getting traffic for ‘cannot download pdb i-tasser’ so maybe it is not just me. I dropped them an email, thanks. I will let you know how it turns out.