AMBER Archive (2009)

Subject: Re: [AMBER] AMBER 94 Charges for terminal LYN, GLH, ASH and CYM residues and some other unusual cases.

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Sun May 10 2009 - 00:49:55 CDT

Dear Tod,

> But I think the question is still there even with these later force fields.
> Also, I think the ff99SB still uses the 94 charges.

Yes, you are right the charges comes from the Cornell et al. force
field (ff94). The force fields that showed up after ff94 (i.e. ff96,
ff98, ff99 or even ff99SB) derives all from the Cornell et al. force
field. This is in ff94 that were designed the charges as you said, but
also the force field atom types, the van der Waals parameters etc...
Usually, these successive 'adapatations' (except for ff99) only
contain several new dihedrals which were refited. Only the Duan et al.
force field proposed a new set of charge values; using however the
same rules for the force field atom type assignement.

> I am writing some code to assign AMBER 94 charges to proteins.
>
> I am using the data in the amino94.in, aminoct94.in and aminont94. files
> from the amber10.ffparms download.

We have worked on Amber & GLYCAM force field conversion, and also used
those *.in force field topology databases as reference for all the
Amber force fields. See AmberFFC
http://q4md-forcefieldtools.org/AmberFFC/

> These files contain 3 sets of charge data for LYS: non-terminal residue
> form, amino-terminal (NH3+ end group) residue form and carboxyl-terminal
> (COO- end group) residue form).
>
> However, only data for the non-terminal residue form of LYN (the
> deprotonated form of Lysine, where the R group is neutral) appears to be
> present.

You are right. Some fragments were not generated because not that
often used I guess.

> Aplogies if this data exists and I'm not finding it. But if not, what
> is
> the recommended procedure to deal with charges for terminal LYN
> residues?

You need to generate these "missing" (central) H2N-terminal and
HOOC-terminal fragments. The reference in this type of work is:
Cieplak et al. Application of the multimolecule and
multiconformational RESP methodology to biopolymers: Charge derivation
for DNA, RNA, and proteins. J. Comput. Chem. 1995, 16, 1357-1377.
However, this reference only described the charged terminal residues...

You can also look in R.E.DD.B. (closed until Monday) where you will
find many different types of molecular fragments (you could follow as
examples), and use R.E.D. or R.E.D. Server to generate your own
fragments; See http://q4md-forcefieldtools.org/.
You could also follow your own strategy, and by providing the required
P2N input files for LYN, GLH, ASH and CYM you will generate what you
need in a single R.E.D. job.

> Similarly, data for the terminal forms of GLH, ASH and CYM appears to be
> absent.

Yes

> Lastly, does charge data exist for the neutral end group forms of the
> amino-terminal and carboxy-terminal residues?

I do not know. You will find examples of such terminal fragments in R.E.DD.B.

regards, Francois

> On Sat, 09 May 2009 14:07:08 -0400, Adrian Roitberg <roitberg_at_qtp.ufl.edu>
> wrote:
>
>> Dear Todd,
>> I am sure more people will tel you this, but please, please please do
>> not use Amber 94 parameters !
>> They are not recomended anymore, and we have better stuff out there.
>>
>> Try instead ff99SB or ff03.
>> Adrian
>>
>> todd_at_semichem.com wrote:
>>
>>> Hello,
>>>
>>> I am writing some code to assign AMBER 94 charges to proteins.
>>>
>>> I am using the data in the amino94.in, aminoct94.in and aminont94. files
>>> from the amber10.ffparms download.
>>>
>>> These files contain 3 sets of charge data for LYS: non-terminal residue
>>> form, amino-terminal (NH3+ end group) residue form and carboxyl-terminal
>>> (COO- end group) residue form).
>>>
>>> However, only data for the non-terminal residue form of LYN (the
>>> deprotonated form of Lysine, where the R group is neutral) appears to be
>>> present.
>>>
>>> Aplogies if this data exists and I'm not finding it. But if not, what
>>> is
>>> the recommended procedure to deal with charges for terminal LYN
>>> residues?
>>>
>>> Similarly, data for the terminal forms of GLH, ASH and CYM appears to be
>>> absent.
>>>
>>> Lastly, does charge data exist for the neutral end group forms of the
>>> amino-terminal and carboxy-terminal residues?
>>>
>>> Thanks!
>>> Todd

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber