Check the format of the pdb - working with vms isn't a good measure of a
compliant pdb file.
Also, try the smaller pdb with leap, not cpptraj. Try to truncate at a
location including the entire last residue.
On Mon, Nov 25, 2024, 5:50 PM Christian Seitz <cgseitz.uchicago.edu> wrote:
> Hi Carlos,
>
> Thanks for your helpful suggestions, which narrowed down the issue. I made
> a truncated pdb containing the first 100 atoms and loaded that into cpptraj
> to create a small output parm7 file. This file wouldn't load into parmed,
> saying the parm7 was an incorrect file type. When I load it into cpptraj to
> check the 'parminfo' command, many system components such as bonds, angles,
> and dihedrals are missing:
> > parminfo
> [parminfo]
> Topology test.parm7 contains 100 atoms.
> Original filename: test.parm7
> 100 residues.
> 100 molecules.
> Box: Orthogonal
> >
> The original, 3 million atom system is also missing molecules, in addition
> to bonds, angles and dihedrals:
> > parminfo
> [parminfo]
> Topology protein.parm7 contains 3276189 atoms.
> Original filename: protein.parm7
> 2252684 residues.
> 0 molecules.
> 10 bonds (0 to H, 10 other).
> Box: Orthogonal
> >
> However, if I load the original, 3 million atom system into VMD as a pdb,
> it looks fine - there are no visibly odd angles and the solvent appears
> normal. How would you recommend creating the missing molecule and bond
> information?
>
> Best,
> Christian
>
> On Fri, Nov 22, 2024 at 1:45 PM Carlos Simmerling <
> carlos.simmerling.gmail.com> wrote:
>
>> Hi Christian, I don't have any solutions for you, but I can make a few
>> suggestions:
>>
>> 1) use the parmed program on your prmtop and use the "printdetails"
>> command to look a some parts (for example, printdetails :1-10). Does it
>> look reasonable?
>> 2) make sure to look at your leap.log file when it loaded your PDB. There
>> will be a lot of useful info, especially about things that leap did not
>> recognize or did not match the libraries (for example, n-terminal MET does
>> not have an atom named HT1). If you don't have access to the log (because
>> of using CHARMM-GUI), then see the parmed suggestion above.
>> 3) make a new PDB file that is much smaller as a test. Try keeping only
>> the first 100 or so atoms, and then process that one. If may be easier to
>> see what is happening
>> 4) keep in mind the limits of standard PDB format, especially for very
>> large systems. Some programs will move the columns around once you have
>> more atoms than can fit, but that isn't allowed by PDB standard and I don't
>> know if CHARMM-GUI can process a PDB file with these changes.
>>
>> carlos
>>
>> On Fri, Nov 22, 2024 at 4:36 PM Christian Seitz via AMBER <
>> amber.ambermd.org> wrote:
>>
>>> Hello,
>>>
>>> I am trying to minimize a 3 million atom system in AMBER20 that was
>>> created
>>> through CHARMM-GUI. The CHARMM-GUI outputs had some issues and stopped
>>> numbering the atoms after a certain point, necessitating some manual
>>> edits
>>> and thus a potential source of error. I renumbered the atoms in my pdb
>>> file, and converted to parm7/rst7. When I try an initial minimization, I
>>> get this error, also shown in the attached error log:
>>>
>>> | ERROR: PMEMD does not support single H residues!
>>> | Atom number 1, Residue number 1
>>>
>>> However, my system is not meant to contain any single H residues and
>>> contains no dummy atoms. The pdb file is formatted like this (looks
>>> normal
>>> on terminal, :
>>> CRYST1 307.624 306.579 415.299 90.00 90.00 90.00 P 1 1
>>> ATOM 1 N MET X 1 -7.171 -15.622-119.707 1.00 0.00
>>> PROA
>>> ATOM 2 HT1 MET X 1 -6.809 -14.657-119.834 1.00 0.00
>>> PROA
>>>
>>> while the structure summary of the parm7 file is this:
>>> %FLAG POINTERS
>>> %FORMAT(10I8)
>>> 3276189 0 0 10 0 0 0 0 0
>>> 0
>>> 3276189 2252684 10 0 0 1 0 0 0
>>> 0
>>> 0 0 0 0 0 0 0 1 2854
>>> 0
>>> 0
>>>
>>> These files are 100-300 MB each but I can send Google drive links to them
>>> if necessary. Do you have any ideas or suggestions on what the error may
>>> be? Thanks in advance!
>>>
>>> Best,
>>> Christian
>>> --
>>> *Christian Seitz*
>>> PhD, Dept. of Computer Science | UChicago
>>> cgseitz.uchicago.edu
>>> [image: www.linkedin.com/in/christianseitz21
>>> <https://urldefense.com/v3/__http://www.linkedin.com/in/christianseitz21__;!!BpyFHLRN4TMTrA!7U_MIM5KSsvQz3Umf23nrvWFiInBfErDCpahMlFe4PDyxD9pjEfGW0NmK2vutmodeEP7ZmKS7AsHOVBClrmnq15qP19Q_Q$>
>>> ]
>>> <http://www.linkedin.com/in/christianseitz21
>>> <https://urldefense.com/v3/__http://www.linkedin.com/in/christianseitz21__;!!BpyFHLRN4TMTrA!7U_MIM5KSsvQz3Umf23nrvWFiInBfErDCpahMlFe4PDyxD9pjEfGW0NmK2vutmodeEP7ZmKS7AsHOVBClrmnq15qP19Q_Q$>
>>> >
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>> <https://urldefense.com/v3/__http://lists.ambermd.org/mailman/listinfo/amber__;!!BpyFHLRN4TMTrA!7U_MIM5KSsvQz3Umf23nrvWFiInBfErDCpahMlFe4PDyxD9pjEfGW0NmK2vutmodeEP7ZmKS7AsHOVBClrmnq17BaP0MLg$>
>>>
>>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 25 2024 - 15:00:02 PST
