Hi Clarie,
Unfortunately, now MCPB.py does not support the recognition of nucleotides automatically. The way to solve your problem is to provide a mol2 file for the nucleotides by yourself and provide it through the naa_mol2files variable in the MCPB.py input file, along with providing an relevant parameter file through the frcmod_files variable (this file is not needed if the force field you chose already includes the parameters for nucleotides) in the MCPB.py input file.
Because Mg2+ is pretty ionic, so usually a nonbonded model is enough. Does the Mg2+ ions leave the original sites if you use a nonbonded model? Some ionic bonds are pretty strong and it is not easy for the ions to leave its original sites.
Pengfei
> On Sep 2, 2020, at 5:43 AM, Claire Claire <anyron.gmail.com> wrote:
>
> Dear all,
>
> I am currently trying to run an MD simulation of an RNA molecule stabilized
> by several Mg2+ ions using the AMBER16 package.
> I need to parametrize the metal coordination centers in order to avoid the
> Mg2+ leaving their original sites. I tried MCPB as I already used it for
> metalloenzymes. However, the system the first MCPB step created does not
> cap the nucleotides as it would do with the amino acids in order to do the
> QM calculations (Gaussian) as it leaves opened valences. I decided to
> manually cap the nucleotides and generate Gaussian input files so that the
> charge calculation would be correct, but now that I want to run the second
> MCPB step I’m facing a discrepancy between mi QM outputs and the
> *standard.fingerprint files produced in the first step. And if I try to run
> the first MCPB step using PDBs with the caped nucleotides, the program
> doesn’t seem to understand the molecules I’m giving it.
> Can anyone give me a bit of insight on how to tackle this issue? Is there
> another version of MCPB that allows the parametrization I’m trying to do?
>
> Claire
> _______________________________________________
> 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 Thu Sep 10 2020 - 14:00:03 PDT
