Re: [AMBER] Missing mtkpp directory after installation of AmberTools21

From: Anthony Nash <>
Date: Fri, 23 Jul 2021 09:26:25 +0000

Hi Pengfei,

Thank you for the information.

I've made progress with the tutorial. The example presents a single Zinc protein complex, however, I'm dealing with a protein that holds x2 Zinc and x4 Calcium ions. I've successfully reached the second step; " -i -s 1 (from the example)", however, I am struggling to understand how I can make this step work for a protein with more than one ion.

My input file for looks like this:

original_pdb 4auo_enzyme_fixed_H.pdb
group_name MMP1
cut_off 2.8
ion_ids 5832 5833 5834 5835 5836 5837
ion_mol2files calcium_atom_1101.mol2 calcium_atom_1102.mol2 calcium_atom_1103.mol2 calcium_atom_1104.mol2 zinc_atom_1201.mol2 zinc_atom_1202.mol2
large_opt 1
force_field ff99SB

However, the process only generates one set of large, small and standard models for GAMESS/GAUSSIAN.

Please suggest how I can make this step compatible with systems that have more than one ion? Having (a) read ahead in the tutorial, and (b) performed all the maths and code to do this for Amber force fields myself several years ago, I am struggling to visualise how can set up the necessary files for multiple ions. Surely, the single set of files this generates (which will have 6 metal ion centres) can't be right? I suspect I have misunderstood something.

Many thanks

Kind regards
Dr Anthony Nash PhD MRSC

Senior Research Scientist
Nuffield Department of Clinical Neurosciences
RMCR Kellogg College
University of Oxford

From: Pengfei Li <>
Sent: 23 July 2021 03:24
To: AMBER Mailing List <>
Subject: Re: [AMBER] Missing mtkpp directory after installation of AmberTools21

Hi Anthony,

> I'm attempting to parameterise 4 calcium ions and 2 zinc ions in a metalloprotease. The last time I parameterised metal for a protein was quite a few years ago and I did it by writing my own 2nd derivative of the Hessian-based model using a Gaussian log file. Unfortunately, I don't have access to a Gaussian license anymore. I've requested a GAMESS-US license, but I've never used that before.

For the Ca2+ ions, a nonbonded model would be a suitable choice. For Zn2+ ions, you can use the bonded model. You can check this webpage for more information about the difference between the bonded model and nonbonded model: <>.

> Are there any empirical calculations I can perform to derive force constants? To satisfy charges I believe I can still use the RED server.

Yes, for Zn2+, there is an option 2e for step 2, which corresponds to an empirical method to derive the force constants for Zn2+ containing systems: <>.

Hope it helps,
AMBER mailing list
AMBER mailing list
Received on Fri Jul 23 2021 - 02:30:02 PDT
Custom Search