Sasha,
I'd suggest to create actual parameters for modified residues, i.e.
model the two peptides as two chains, and create parameters for the
linking moiety: either treat it as modified cysteine residue with
branching ability like CYX or, depending on the chemical nature of your
linker, reuse the CYX residue and create an additional linking residue.
Of course, you need to generate suitable parameters and atomic charges
for your system. But you had a full MM description then and not just an
harmonic constraint between two molecules. However, it also depends on
your actual research question.
We used a somehow similar approach before (see
10.1021/acs.jmedchem.1c01574; supplement p.22f). If you need more
specific support for your system, feel free to contact me off-list.
Best,
Anselm
Bioinformatik | NHR.FAU
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)
Germany
Am 26.02.2024 um 05:27 schrieb Sasha Buzko via AMBER:
> Dear Amber users,
> I'm trying to model a system that consists of a protein, a non-peptide linker covalently bound to a cysteine and a short peptide covalently bound to the other end of the linker.
> Since this system is effectively branched, I'm trying to figure out how to set it up.
>
> I've tried treating the protein and the linked peptide as two separate chains and using restraints on the cysteine and the bonded functional group of the linker to hold them together. For example:
> Restraint group 1 # During the heating phase, restraining all protein atoms
> 100.0
> RES 1 29 31 618
> END
> Restraint group 2 # The restrained residues in question have a higher value that is used everywhere
> 1000.0
> RES 30 30 619 619
> END
> END
>
> The restrained linker keeps drifting off sooner or later. It may hold in place during most of the equilibration but when it hits production the linked structure floats away eventually. It happens both in explicit (TIP3P, OPC or TIP4PEW) and implicit solvent. I am setting the ntr=1 flag everywhere.
>
> I've also attempted to use the ibelly option but this one is not usable with the GPU implementation and even in many cases on CPU.
>
> Finally, I'm trying to define the entire linked structure as a non-standard amino acid and run it through antechamber. But this is predictably taking forever and is not going to be practical even if it works. Besides, not sure about the accuracy of the charges for amino acids that will come out of it.
>
> Has anyone had to deal with something like this? Am I missing something or is this the limit of the restraint function? Maybe there is a better way to approach this?
>
> Thank you in advance for any suggestions.
>
> Sasha
>
>
>
>
> ---
> Oleksandr "Sasha" Buzko
> Computational Biology
> ImmunityBio
> 9922 Jefferson Blvd
> Culver City, CA 90230
>
> CONFIDENTIALITY NOTICE
> This e-mail message and any attachments are only for the use of the intended recipient and may contain information that is privileged, confidential or exempt from disclosure under applicable law. If you are not the intended recipient, any disclosure, distribution or other use of this e-mail message or attachments is prohibited. If you have received this e-mail message in error, please delete and notify the sender immediately. Thank you.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 26 2024 - 03:00:02 PST
