RE: Martini CG models parameters errors

From: Mahad Gatti (
Date: Wed Sep 28 2016 - 05:40:51 CDT

Dear Fotis and Josh,

many thanks for the help. I was able to solve the problem by adding the "martini_v2.2.namd.prm". Now I'm ready to go with some CG simulations.

Best regards,

Da: []
Inviato: lunedý 26 settembre 2016 21.58
A:; Mahad Gatti
Oggetto: Re: namd-l: Martini CG models parameters errors

Hello Mahad,

You have to load BOTH of the required prm files containing Martini
parameters in NAMD format, namely, the following:
(this contains all particle definitions and non-bonded terms, as well as
most of the bonded terms for lipids and proteins).
(this contains terms for all bonds that were defined as "constraints" in
GROMACS - since NAMD has no LINCS constraints, these are modeled as
normal bonded terms, but with very high force constant values, which is
the main problem of the NAMD implementation).

Both of these files are in the "toppar" directory (not the one in the
GROMACS output). Note that in order for the angles to work correctly,
you need to use the files with "namd" in their name, the others are for
CHARMM (the program, not the FF) and do not contain the cosine angle terms.

Having said the above, I would advise caution when using NAMD and
MARTINI. Because of this trick they used with the constraints, anything
with a timestep of 10 fs or higher regularly leads to crashes when
proteins are included (lipids, with the exception of glycolipids,
cholesterol and other similar structures are fine, since they do not use
constraints). Also, if you chose the polarizable water model when
building your system, you can pretty much forget about using a standard
NAMD buid, since it doesn't implement virtual sites. There is a
workaround for that, but it requires editing the source code.

Hope I helped,

On 09/26/2016 06:48 PM, Vermaas, Joshua wrote:
> Hi Mahad,
> I don't recognize the P5C atomtype, as its not standard CHARMM. I'm
> guessing its a martini atomtype, which means that there should be a P5C
> term in the nonbonded section of the parameter file you have still
> loaded. If there isn't one, that is your problem. :) If you look at the
> psf/pdb pair in VMD, what atoms have the P5C type? It could be that it
> represents a non-protein component, which probably wouldn't be the the
> amino acids file.
> -Josh
> On 09/26/2016 03:01 AM, Mahad Gatti wrote:
>> Hi Josh,
>> many thanks for the answer. I did like you said and I kept only the:
>> parameters toppar/
>> while commenting the line:
>> #parameters ../toppar_water_ions_namd.str
>> but now I get this error:
>> Am I (again) loading the wrong parameters?
>> Best regards,
>> Mahad
>> ________________________________________
>> Da: Vermaas, Joshua []
>> Inviato: giovedý 22 settembre 2016 17.36
>> A:; Mahad Gatti
>> Oggetto: Re: namd-l: Martini CG models parameters errors
>> Below is the offending line. OC is an atomtype in toppar_water_ions, but
>> only as an NBFIX term. The file doesn't provide VDW information for that
>> atomtype though (normally I think the protein or lipid file handles
>> that), and so you get the error. If you are using martini, why would you
>> need TIP3 water parameters anyway? Just comment out the line, as I
>> suspect the martini parameter file will give you the parameters you need
>> for the coarse grained water I hope you are using.
>> -Josh
>> On 09/22/2016 09:14 AM, Mahad Gatti wrote:
>>> parameters ../toppar_water_ions_namd.str

Fotis A. Baltoumas
Phd Candidate, Bioinformatics Postgraduate Programme
Department of Cell Biology and Biophysics
Faculty of Biology, University of Athens
Panepistimiopolis, Athens 157 01, GREECE
email :

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:22:30 CST