Previous Thread
Next Thread
Print Thread
Joined: Jan 2021
Posts: 2
S
sdg Offline OP
Forum Member
OP Offline
Forum Member
S
Joined: Jan 2021
Posts: 2
Hi all

I have created coordinate and psf files for my system comprising of a modified nucleic acid inside a protein, embedded in a lipid (step5_input.pdb/psf/crd) using CHARMM-GUI. However, I would like to add one triple bond between atom# 22424 and 23769 and also obtain the updated PSF and CRD files. The bond is supposed to be created between C5M of THY 17 and C1 of DL 61 which is the modification created using CGenff. My understanding is that one cannot use just CHARMM-GUI to create non-covalent bond between a biomolecular and a non-biomolecular part. Is that true? After I get the output from CHARMM-GUI I dont see any bond between those two although (absent from step5_input.psf) I made sure that CONECT records were there in the input PDB. Now, with this PDD+PSF I want to incorporate the triple bond, update the angle, dihedral etc info in a new PSF and CRD files and submit them to CHARMM-GUI FF-Converter for creating my system correctly. I have limited experience with PSFGEN tools and that's why I am asking for your help. If you can provide me with some help on how to write a sample psfgen script with my system for the triple covalent bond that I am looking for that would be of tremendous help. I am not sure what my PATCH statement would be for PSFGEN to work. You may find the associated files in the attached link. The FF parameters are in the toppar directory.


https://drive.google.com/drive/folders/1zY_zfwIzDgNGJiqxdGAaWSc40un6m7V2

I would be grateful if you share your thoughts on this approach of modelling a modified nucleic acid using CHARMM-GUI. If CHARMM-GUI is not adequate I need to use updated versions of my PSF/CRD files and I would like to have your help with PSFGEN or some other way.



Thank you !

Soumadwip Ghosh

Joined: Sep 2003
Posts: 4,840
Likes: 3
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,840
Likes: 3
Any changes to the topology of your system (typically new covalent bonds) have to be introduced in the residue topology file (RTF), either in a new residue or as a patch (a patch is a special kind of residue which is used to change an existing residue, eg by adding/deleting a bond or atom).

You do not tell us what your modification is (and I cannot tell you how sad I am that I cannot help you with atoms 22424 and 23769 ). CHARMM does have a fairly comprehensive set of modified nucleotides in its force field, perhaps your modification has already been defined? If not, you have some work to do. You mention CGenFF, but it is not clear why you could not use its output.

As far as CHARMM is concerned a triple bond is just a bond. A non-covalent bond is not a bond, and is only treated using electrostatic and vdW interactions.

I am not very familiar with CHARMM-GUI, but my understanding is that it cannot do anything which is not already present in the CHARMM program or the CHARMM force field.

Please discuss with your supervisor!


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Joined: Jan 2021
Posts: 2
S
sdg Offline OP
Forum Member
OP Offline
Forum Member
S
Joined: Jan 2021
Posts: 2
Hi Lennart,

I have created the FF parameters for the modification using CGenFF as a part of CHARMM-GUI which builds a full system for simulations. In my input PDB there were 60 DNA residues and one modification (that has four glutamate like moieties) that are supposed to be triple bonded. My question was is CHARMM GUI adequate the create the bond between the nucleic and the non-nucleic parts? I supplied the mol2 file for the modified residue and CHARMM-GUI gave me its FF parameters. However, I don't see the triple bond between the two residues of interest in the output PSF created by CHARMM-GUI. I was wondering if one can specify this bond in CHARMM-GUI or may I take the output from CHARMM-GUI and somehow insert it back, build a complete PSF and start CHARMM-GUI FF-converter again. If there is a way to build everything using CHARMM-GUI for this kind of system one does not need to do things repeatedly.

Joined: Sep 2003
Posts: 8,602
Likes: 21
rmv Online Content
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 8,602
Likes: 21
Questions about CHARMM-GUI functionality should probably be asked via the Contact page on their site.


Rick Venable
computational chemist


Moderated by  lennart, rmv, wonpil 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5
(Release build 20201027)
Responsive Width:

PHP: 7.3.31-1~deb10u1 Page Time: 0.009s Queries: 22 (0.006s) Memory: 0.7434 MB (Peak: 0.7951 MB) Data Comp: Off Server Time: 2022-08-18 20:21:37 UTC
Valid HTML 5 and Valid CSS