[phenixbb] secondary structure presentation in VMD

[Tristan Croll]

Since I'm not actually all that familiar with PyMol scripting, here's 
how you can write out secondary structure annotations from ChimeraX and 
apply them in VMD for an arbitrarily-sized structure. First, open the 
model in ChimeraX and go to the Python console (Tools/General/Shell). 
Then copy/paste the below:

def write_secondary_structure(model, filename):
     from chimerax.core.atomic import Residue
     r = m.atoms.residues
     ss_types = r.ss_types
     type_dict = {
         Residue.SS_COIL:    'C',
         Residue.SS_HELIX:   'H',
         Residue.SS_STRAND:  'E'
         }
     with open(filename, 'w') as f:
         print(','.join((type_dict[s] for s in r.ss_types)), file=f)

from chimerax.core.atomic import AtomicStructure
m = session.models.list(type=AtomicStructure)[0]
write_secondary_structure(m, 'structure.txt')

This will write out the secondary structure information to structure.txt 
using VMD's annotation format. Then, open your model in VMD, open the 
console (Extensions/Tk Console) and copy/paste the below:


proc apply_secondary_structure {model infile} {
     set infile [open $infile r]
     set fdata [read $infile]; puts "Read structure info"
     set f_structure [split [lindex $[split $fdata "\n"] 0] ","]; puts 
"Successfully split data"
     $model set structure $f_structure; puts "Applied structure to model"
     close $infile
}

set model [atomselect top all]
apply_secondary_structure $model structure.txt
$model delete

... and your cartoon should now look as expected.


On 2018-01-21 18:23, Tristan Croll wrote:
> I’m guessing this structure is on the large side? Unless something has
> changed since I last used it, VMD ignores secondary structure
> annotations in the PDB file. It calculates them instead using an old
> version of STRIDE, which fails beyond 9,999 residues (or possibly
> atoms? Can’t remember now). One can hack together a quick-fix solution
> fairly easily (basically split the structure into smaller pieces,
> calculate for each of those then copy the results back to the master
> structure). If you really want the secondary structure in VMD to match
> what Pymol sees, you could do it fairly easily with a little scripting
> in both packages. VMD gets/sets secondary structure as a simple Tcl
> array of characters, one per residue. So all you’d need to do in Pymol
> is a simple loop over all residues to get their secondary structure
> designation and write the corresponding character to a text file. Then
> load the text file in VMD, convert the string to a character array,
> and apply it to your model. A mite fiddly, but doable.
> 
> Alternatively, have you tried ChimeraX? It’s really rather nice in
> what it can do.
> 
> Hope this helps,
> 
> Tristan
> 
> 
> 
> 
> 
> Tristan Croll
> Research Fellow
> Cambridge Institute for Medical Research
> University of Cambridge CB2 0XY
> 
> 
> 
> 
>> On 21 Jan 2018, at 15:44, Paul Emsley <pemsley at mrc-lmb.cam.ac.uk> 
>> wrote:
>> 
>> 
>> Hi Charles,
>> 
>>> On 21/01/2018 15:04, CPMAS Chen wrote:
>>> 
>>> This may be a little off-topic. I have posted this in VMD users mail 
>>> list. I want to try here too to see
>>> if anyone experienced similar issue.
>>> VMD and pymol gave different presentation on the secondary structure 
>>> of a protein.
>>> I correctly labelled all helices and sheets.
>> 
>> I wonder what that means. How did you label all helices and sheets and 
>> how did you know that you did it correctly?
>> 
>> 
>> Paul
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb
>> Unsubscribe: phenixbb-leave at phenix-online.org
> 
> 
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org