Dear Matt,
If you use Antechamber you should check:
- if the total charge of the whole molecule is correct: answer 'Not exactly'
- if chemical equivalencing is done: answer NO
- if constraints used during the fit are correctly applied: answer
'Not always'
- if the total charge of a molecular fragment is what you expect:
answer 'likely No'
being aware that Antechamber does neither handle multiple conformation
charge fit nor multiple molecule charge fit
All that to support what Aron reported previously: I would like to
suggest you to use the R.E.D. III.x tools or R.E.D. Server.
We will release very soon the R.E.D. III.5 tools...
Finally, you might look at FF library file formats different from 'prep'...
See
http://q4md-forcefieldtools.org/Tutorial/leap-mol3.php
R.E.D. and R.E.D. Server use the mol2 file format...
regards, Francois
Quoting Aron Broom <broomsday.gmail.com>:
> I wouldn't use antechamber to make a non-standard residue, but maybe that's
> just me.
>
> when you edit your residue in leap does it look ok? That is, are all the
> bonds actually present that you wanted?
>
> I have no idea about this OMIT_NAME.
>
> I suggest making a dipeptide of your residue. Load the structure in leap,
> edit it and add all the bonds manually. Edit the charges and enter
> something reasonable/similar to what you got using Antechamber, then relax
> the structure. Then save that as a pdb. Submit that to the RED Server
> using their automated non-standard residue generating option. You are
> done. At least that is what I would do.
>
> ~Aron
>
> On Mon, Feb 13, 2012 at 11:55 PM, Matthew D Antalek <mantale1.binghamton.edu
>> wrote:
>
>> Dear all,
>>
>> I was able to get through most of the tutorial
>> http://ambermd.org/antechamber/pro4.htm<
>> http://ambermd.org/antechamber/pro4.html>
>> without
>> any problems except for then end. When I created the prepi file, I got this
>> message:
>>
>>
>> Info: there is a bond linking a non-head and non-tail residue atom (C1) and
>> an omitted atom (H3).
>> You need to specifically add this bond in leap using the command
>> 'bond <atom1> <atom2> [order]'
>> to link C1 to an atom in another residue (similar to disulfide bonds)!
>>
>> Info: there is a bond linking a non-head and non-tail residue atom (C1) and
>> an omitted atom (C3).
>> You need to specifically add this bond in leap using the command
>> 'bond <atom1> <atom2> [order]'
>> to link C1 to an atom in another residue (similar to disulfide bonds)!
>>
>> I am not sure exactly what I should do to correct this ( or if I even need
>> to ). I basically copied what the tutorial did, naming all the atoms
>> besides the 3 mainchain atoms as OMIT_NAME. Also, when did a quick check of
>> the structure in xleap (creating a polypeptide NALA 3FY CALA), I was able
>> to create a sequence using the new amino acid, but it seems that all the
>> atoms I listed as OMIT_NAME atoms are not displaying.
>>
>> Thanks,
>> Matt
>>
>>
>> On Sun, Feb 12, 2012 at 11:00 PM, Matthew D Antalek <
>> mantale1.binghamton.edu
>> > wrote:
>>
>> > Thanks, I did not know about the RED server before. I will try to go
>> > through with the given tutorial and also look at the one on the RED
>> server.
>> >
>> > Thanks,
>> > Matt
>> >
>> >
>> > On Sun, Feb 12, 2012 at 10:50 PM, Aron Broom <broomsday.gmail.com>
>> wrote:
>> >
>> >> Just as another tidbit of information, the RED server has a tutorial for
>> >> this kind of thing. The problem you are likely coming across, in which
>> >> you
>> >> say it is not recognized as an amino acid, is that it doesn't have
>> connect
>> >> information. You need to add a connect0 and connect1, probably also a
>> >> head
>> >> and tail definition. such that it knows your molecule is not a
>> stand-alone
>> >> molecule, but intended to be used as a fragment. This information tell
>> >> LeAP which atoms in your residue are meant to connect to other residues
>> >> when they are a chain. Again I recommend the RED server and associated
>> >> tutorial, they have a special option on their server for making
>> >> non-standard amino aicds.
>> >>
>> >> ~Aron
>> >>
>> >> On Sun, Feb 12, 2012 at 10:40 PM, Jason Swails <jason.swails.gmail.com
>> >> >wrote:
>> >>
>> >> > What do you mean "recognized as an amino acid"? There are very few
>> >> > situations in the amber programs that actually make use of that
>> >> distinction.
>> >> >
>> >> > Also, if it's a modified amino acid I don't know that gaff is the best
>> >> > force field to be using to parametrize it. Does FF10 not have the
>> >> > appropriate parameters? Has anyone else parametrized that residue
>> yet?
>> >> > (RED Database is a good place to look, as is the Bryce parameter
>> website
>> >> > that is easily found googling bryce and amber).
>> >> >
>> >> > In any case, you should run some short dynamics on some test systems
>> >> with
>> >> > your amino acid residue (maybe sandwiched between a couple alanines or
>> >> > something) and just make sure that some of the properties you can
>> >> measure
>> >> > make physical sense (like phi-psi distributions, etc), just to be sure
>> >> that
>> >> > nothing is drastically wrong.
>> >> >
>> >> > HTH,
>> >> > Jason
>> >> >
>> >> > --
>> >> > Jason M. Swails
>> >> > Quantum Theory Project,
>> >> > University of Florida
>> >> > Ph.D. Candidate
>> >> > 352-392-4032
>> >> >
>> >> > On Feb 12, 2012, at 9:41 PM, Matthew D Antalek <
>> mantale1.binghamton.edu
>> >> >
>> >> > wrote:
>> >> >
>> >> > > Hi all,
>> >> > >
>> >> > >
>> >> > > I am having a bit of trouble with getting AMBER to recognize the
>> >> > > topology file for a non standard amino acid residue I have created.
>> I
>> >> was
>> >> > > able to create the parameter-topology file (.prmtop) but I do not
>> >> > > understand how to create some other files so that the system can
>> >> > recognize
>> >> > > the residue as an amino acid residue. I have tried to work with this
>> >> > > tutorial : http://ambermd.org/antechamber/pro4.html but I do not
>> have
>> >> > > access to gaussian. I was able to generate the parameters for the
>> >> > molecule
>> >> > > using GAFF, so if anyone knows how to get the system to register the
>> >> > > molecule as an amino acid that would be great.
>> >> > >
>> >> > > Thanks,
>> >> > > Matt
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 14 2012 - 08:30:02 PST
