VMD-L Mailing List

From: Luis Gracia (lgraval_at_gmail.com)
Date: Wed Mar 21 2007 - 11:12:32 CDT

Hi! You might want to try the iTrajComp plugin. It can calculate a
contact map by residue and graph it in a matrix-like display. Clicking
in the matrix will display the residues in vmd. The output can also be
save in plotmtv, ps and text (tabular and matrix formats). Let me know
if you are interested, I'll be happy to give you a copy.

http://physiology.med.cornell.edu/faculty/hweinstein/vmdplugins/itrajcomp/
(the contact map you want is done using the 'dist' option, no the
'contacts' one)

Best,

Luis

John Stone said the following on 03/21/2007 02:42 AM:
> Hi,
> I'm forwarding your note to the author of the Contact Map plugin
> in case he has suggestions on this. I believe I'm the one that
> added the text comment to the source file about the buggy behavior,
> and that this is why that functionality is not exposed to the graphical
> interface, as I recall. Unless Barry (the author) has time to look
> into this in the short-term, you might be better off using his script
> as a "guide" and write your own implementation. I think a version that
> just outputs the matrix to text would be far easier to write/debug than
> the graphical version that Barry wrote, so this would be one reason to
> develop your own script. It would likely be faster as well, and since
> this is an O(N^2) type of computation (or O(NlogN) if you use an acceleration
> data structure), I suspect you'd be much better off doing your own version
> if you have big structures to process.
>
> John Stone
> vmd_at_ks.uiuc.edu
>
> On Tue, Mar 20, 2007 at 10:18:03PM +0800, Guo Zhi wrote:
>> Hi,
>> I wish to output the residue distance matrix data, to save it
>> in the format of numerical data in a text file rather than the default
>> snapshot function in contactmap plugin. I found a function called
>> writeDataFile() in the source, but as its comment says, it can not
>> work normally, always missing some data when I tried to use it. Does
>> anyone have got a better solution? Thanks.
>>
>> Guo Zhi
>