VMD-L Mailing List

From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Fri Apr 16 2021 - 10:30:35 CDT

HI Ashar
Thanks for your suggestions. As I wrote, I was using the test version 1.9.4
of VMD. In doubt, I have now tried the stable 1.9.3 and psf/pdb were
generated, no vecsub error.

I still have to check the psf/pdb but I hurried to avoid other people
loosing time. I'll not be back if psf/pdb are correct.

all the best
francesco

On Fri, Apr 16, 2021 at 12:31 PM Ashar Malik <asharjm_at_gmail.com> wrote:

> Hi Francesco,
>
> The error
>
> vecsub: two vectors don't have the same size
>
> is because the two compared entities $x and $y don't have the same number
> of entries being compared.
> For instance if you do (from
> https://www.ks.uiuc.edu/Research/vmd/vmd-1.7.1/ug/node163.html)
>
> vecsub {10 9.8 7} {0.1 0 -0.1}
>
> you will get
>
> 9.9 9.8 7.1
>
> but if you do
>
> vecsub {10 9.8 7} {}
>
> From what I recall, it will return the above error [vecsub: two vectors
> don't have the same size].
>
> Having said that a wild guess to what is causing your issue might be an
> unbalanced selection.
>
> Here is an example from my work where I encountered this error and how I
> fixed it.
>
> I was faced with this when I had a protein A and a modified version of
> protein A - with the modification being just a slight change in
> conformation. I wanted to compute the backbone deviation between the normal
> and modified forms and I selected the backbone using the selection
>
> "protein and name CA"
>
> in one protein it selected all amino acids (as expected)
>
> in the other protein the above selection picked all amino acids except one
> residue (somewhere in the middle of the chain).
>
> From my two selections I was returning a list of residues and comparing
> each lists's CAs with the other list's - since there was a one to one
> correspondence between the lists.
>
> There was one for loop, as the expected length of both the lists was the
> same, and at the last iteration because one list was shorter compared to
> the other I got the above error.
>
> replacing the above selection
>
> "protein and name CA"
>
> with
>
> "all and name CA"
>
> as I only had 1-protein chain in my loaded molecule fixed the problem.
>
> If possible you can look at your 30A cut and the selection it produces and
> why exactly are you using vecsub and what is vecsub operating on and then
> check where those are coming from - you should be able to identify the
> cause of the error.
>
> Hope the above example is of some help.
> Sorry I couldn't give you an exact answer - but someone else might :)
>
> Regards,
> /Ashar
>
>
> On Fri, Apr 16, 2021 at 5:51 PM Francesco Pietra <chiendarret_at_gmail.com>
> wrote:
>
>> Hi all
>> I am faced by the following error while executing autopsf with a
>> nucleotide-protein system at the Guess/Spl;it Chains level (a 30A sphere
>> cut around the small-molecule ligand)
>> VMD 1.9.4a51
>> vecsub: two vectors don't have the same size
>> vecsub: two vectors don't have the same size
>> while executing
>> "vecsub $x $y"
>> (procedure "vecdist" line 2)
>> invoked from within
>> "vecdist $Ccoords $Ncoords"
>> (procedure "split_protein_and_water_pdb" line 250)
>> invoked from within
>> "split_protein_and_water_pdb "${basename}-temp.pdb""
>> (procedure "::autopsf::aftersels_gui" line 55)
>> invoked from within
>> "::autopsf::aftersels_gui"
>> invoked from within
>> ".autopsf.sels.next invoke"
>> ("uplevel" body line 1)
>> invoked from within
>> "uplevel #0 [list $w invoke]"
>> (procedure "tk::ButtonUp" line 22)
>> invoked from within
>> "tk::ButtonUp .autopsf.sels.next"
>> (command bound to event)
>>
>> I added the parameters for the ligand (which, for the ligand alone worked
>> well, also for MD), and
>> top_all36_cgenff.rtf
>> top_all36_lipid.rtf
>> top_all36_na.rtf
>> top_all36_prot.rtf
>> toppar_water_ions.str
>>
>> Then, with "Everything" selected I clicked on Guess and split chains,
>> getting the above error.
>>
>> I found no errors in the pdb file with VMD Jmol CHIMERA. Histidine was
>> assigned charmm names. Ligand pdb was in accordance with its psf/pdb. The
>> only other hetero was MG.
>>
>> Thanks for any suggestion on what else I could check.
>>
>> francesco pietra
>> I had a similar error in the past for another nucleotide-protein
>> ensemble, but only for a sphere of 60A, while 30A - which went on without
>> errors - was enough. I did not pursue that error further.
>>
>>