AMBER Archive (2009)

Subject: Re: [AMBER] specification of antechamber AC file format?

From: Nicholas Musolino (musolino_at_MIT.EDU)
Date: Wed Apr 15 2009 - 17:07:32 CDT

Thanks very much for your reply, Prof. Case. I know it's been a while
since I first e-mailed the list, but I wanted to follow up.

I've attached a MOL2 file for 1-butanol, which I process with the
parmchk -i orig901.mol2 -f mol2 -o orig901.frcmod

The output I get is below. The zero force parameters are problematic.
> remark goes here
> C.3 0.000 0.000 ATTN, need revision
> H 1.008 0.161 same as hn
> O.3 0.000 0.000 ATTN, need revision
> C.3 0.0000 0.0000 ATTN, need revision
> H 0.6000 0.0157 same as hn
> O.3 0.0000 0.0000 ATTN, need revision

I think I know what the issues could be, but can anyone tell me:
1) Does parmchk require GAFF atomtypes? Or another atomtype format?
2) Does parmchk actually read the bond information in the mol2 file?
3) Do the atom names (field 2) need to be different for each atom,
e.g. H1, H2, ... H10?

More generally, I am able to produce MOL2 files of arbitrary molecules
with (a) SYBYL atom type information and (b) bond information that I
deem trustworthy. What is the best way to obtain an FRCMOD file, and
then a PARM7 file, for later use with NAMD?

Many thanks for any help/information that list members are able to

Best regards,

Nicholas Musolino

Dept. of Chemical Engineering | 617-253-6675 | Room E19-528

On Mar 4, 2009, at 5:39 PM, David A. Case wrote:

> On Tue, Mar 03, 2009, Nicholas Musolino wrote:
>> As a matter of fact, I have MOL2 files with SYBYL atom names and
>> complete bond information, which I generate in an automated way.
>> But no
>> matter what I try, I can't seem to do a straight conversion of this
>> information to .AC format through antechamber. (I also want to add
>> partial charges, but that's not a huge problem.)
> Can you provide an example that fails?
>> Right now I am looking to bypass antechamber and write my own .AC
>> files
>> for use with parmchk.
> It's not clear why you need to write .ac files, which are not really
> intended as a general file format (and it would be good to get rid of
> them). The parmchk program can use mol2 files as input, and you
> seem to
> already have them. Of course, parmchk doesn't know about the sybyl
> force field, so you do need to use antechamber to convert the atom
> types.
> Which returns us to the original question -- what is going wrong with
> Antechamber itself?
> ...dac
> _______________________________________________
> AMBER mailing list

AMBER mailing list