Hi Gustavo,
Are you using a bonded model for the water oxygen? You should try using a
non-bonded model, as i've run into this before, and using nonbonded
approach for the waters worked wonderfully and Pengfei detailed the
solution and cause to this problem in the tutorial (see bottom of the page)
http://ambermd.org/tutorials/advanced/tutorial20/mcpbpy.htm
All the best,
Clorice
On Fri, Nov 4, 2022 at 10:07 PM Gustavo Seabra via AMBER <amber.ambermd.org>
wrote:
> Hi everyone,
>
> I'd really appreciate some help figuring out what I'm doing wrong here.
>
> TLDR: I created a system using MCPB.py, and as soon as MD simulation
> starts, one water molecule collapses with another residue in the metal
> center, and the simulation blows up. I appreciate any help I can get to
> solve this problem. Thanks!
>
> Details:
> I have a metalloprotein in which the Fe(II) center coordinates to 2 HID, 1
> ASP and 1 water, and I used MCPB.y to create the parameters for MD. The
> procedure I followed was:
>
> Starting from an apo-protein prepared in Maestro, the treated with
> pdb4amber,
> $ grep FE apo-protein_pdb4amber.pdb > FE.pdb
> $ grep WAT apo-protein_pdb4amber.pdb > WAT.pdb
> $ metalpdb2mol2.py -i FE.pdb -o FE.mol2 -c 2
>
> # Create parameters for the water:
> $ antechamber -i WAT.pdb -fi pdb -o gaussian_WAT.com -fo gcrt -gv 1 -ge
> WAT.gesp -eq 2 -rn WAT -pf Y
> (Run g09. Note: I use Gaussian09 because the files from Antechamber were
> prepared for it.)
> $ antechamber -i WAT.gesp -fi gesp -o WAT_resp.mol2 -fo mol2 -c resp -eq 2
> -rn WAT -at amber -pf y
> $ ln -s WAT_resp.mol2 ./WAT.mol2
> $ cp $AMBERHOME/dat/leap/parm/frcmod.opc .
>
> # Now, run MCPB.py
> $ MCPB.py -i MCPB.in -s 1 2>&1 | tee mcpb_s1.log
> (Run g16 calculations for the complexes as created by MCPB.py
> Note: I here use Gaussian16, and just alter the SCF algorithm to SCF=YQC,
> which is a newer method suitable for large systems.)
> $ MCPB.py -i MCPB.in -s 2 2>&1 | tee mcpb_s2.log
> $ MCPB.py -i MCPB.in -s 3 2>&1 | tee mcpb_s3.log
> $ MCPB.py -i MCPB.in -s 4 2>&1 | tee mcpb_s4.log
>
> # Modify protein-FE_tleap.in file to
> 1. Use solvateOct with 15A buffer,
> 2. Add `check mol` to get the charges
> 3. Stop after adding waters, so we can calculate the
> number of ions to add.
> 4. Add 'verbosity 2' after sourcing force fields
> 5. Change save commands to save NetCDF files and extensions
> `.ncrst` and `.parm7`
>
> The charge was +7.0, and 24,718 waters were added. So,
> add the following line to protein-FE_tleap.in:
> `addIonsRand mol NA 63 CL 70`
> Now, run it fully to create the system.
>
> # Finally, create the parameters with LEaP
> $ tleap -s -f protein-FE_tleap.in 2>&1 | tee protein-FE_tleap.log
>
> LEaP creates the files with some warnings, but no errors. With those last
> files, I starting an MD protocol where I first run a sequence of
> minimizations where I start with the protein and metal center (including
> the coordinated water) restrained with a constant of 100 kcal/(mol.A)^2,
> then I remove the restraints gradually in a series of minimizations, to
> finally run a small 100 K MD and then start heating the system.
>
> The problem is that as soon as the restraints are weak enough, the system
> collapses. One hydrogen of the coordinated water just sinks into one ASP
> oxygen, and the simulation blows up.
>
> I suppose I must be doing something wrong here, but I really can't figure
> out what is wrong. Does anyone here see what I'm missing? I really
> appreciate any help. I attached here the input I'm using for MCPB.py. I'll
> be happy to provide more information if anyone needs it.
>
> Thanks a lot!
> --
> Gustavo Seabra.
> _______________________________________________
> 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 Tue Nov 08 2022 - 07:00:02 PST
