Lekpa,
> You are right I used the RESP-A1 model and the RRMS results you quote are
> true. The charge in the Mol_m1-o1.mol2 file for the CH3CO fragment is about
> 0.2 which is indeed quite different from zero. Also the charge on the OMe
> oxygen is quite defferent between constrained and non constrained fit:
> -0.4435 for unconstrained agaist -0.1909. Even the hydrogens are also too
> different, 0.0819 and 0.1075 respectively. It seems as if the may not be the
> right constraints and or molecule.
ok, so this is normal that the RMSD of the fit with this intra-mcc is
not that good. Now, what about EtCO-OMe ? Does it improve the fit &
does it at least go in the right direction ?
Corresponding P2N input file:
REMARK INTRA-MCC 0.0 | 1 2 3 4 5 6 7 8 9 | Remove
REMARK
ATOM 1 C1 EST 1 X.XXX Y.YYY Z.ZZZ CX
ATOM 2 H11 EST HX1
ATOM 3 H12 EST HX2
ATOM 4 H13 EST HX3
ATOM 5 C2 EST CY
ATOM 6 H21 EST HY1
ATOM 7 H22 EST HY1
ATOM 8 C3 EST CZ
ATOM 9 O4 EST OZ
ATOM 10 O5 EST O1
ATOM 11 CT6 EST C1
ATOM 12 H6 EST H11
ATOM 13 H6 EST H12
ATOM 14 H6 EST H13
regards, Francois
> On Tue, Mar 2, 2010 at 9:44 PM, FyD <fyd.q4md-forcefieldtools.org> wrote:
>
>> Lekpa,
>>
>> Using RED as you suggested, the the derivation was carried out but it
>>> appears that the constraint really reduced the accuracy of the fit from a
>>> value of 0.11363 to 0.29321. Since I do not have a lot of experience with
>>> charge derivation in this manner I am not particularly sure if this is an
>>> acceptable number. I suppose I will try other compounds to see if I get a
>>> better fit. Any thoughts welcome! Thanks. Lekpa.
>>
>> - When adding intra-mcc(s) in a P2N file, R.E.D. automatically generates
>> two fits: one without this/these intra-mcc (useful as a reference) with the
>> corresponding files generated (among many others): inputX_m1, outputX_m1,
>> punchX_m1 (X = 1 or 2 so far), & the FF library: Mol_m1-o1.mol2 and another
>> one with the intra-mcc(s); files generated: inputX_m1.sm, outputX_m1.sm,
>> punchX_m1.sm & Mol_m1-o1-sm.mol2
>>
>> See
>> http://q4md-forcefieldtools.org/Tutorial/P2N/Central-frag-Pept/listing-1mol.pdf
>>
>> - (I suppose you use the RESP-A1 charge model) if I understand you, you
>> got:
>> ESP relative RMS (SQRT(chipot/ssvpot)) 0.11363 in output2_m1
>> ESP relative RMS (SQRT(chipot/ssvpot)) 0.29321 in output2_m1.sm
>> If so, yes, this is bad. Surprising...
>>
>> - What is the total charge (you have to compute it manually) of this CH3CO
>> group in the fit WITHOUT the intra-mcc (i.e. in Mol_m1-o1.mol2) ? if too
>> different to zero this means that the CH3CO is not the right group to be
>> constrained and MeCO-OMe is not the right model (even if the organic
>> function is correct).
>>
>> I will check that today. I let you know...
>>
>> regards, Francois
>>
>> On Tue, Mar 2, 2010 at 8:03 AM, Lekpa Duukori <duukori.gmail.com> wrote:
>>>
>>> Thanks Francois. I indeed need an OMe for an ester group. I will try to
>>>> follow your suggestions.
>>>>
>>>> Lekpa
>>>>
>>>>
>>>> On Tue, Mar 2, 2010 at 3:29 AM, FyD <fyd.q4md-forcefieldtools.org>
>>>> wrote:
>>>>
>>>> Dear Lekpa,
>>>>>
>>>>>
>>>>> Please does anyone know if there is an OMe fragment in the AMBER FF
>>>>>
>>>>>> parameters? I want to do an OMe cap instread of the usial NMe
>>>>>>
>>>>>> I cannot find one, just checking to see if I somehow missed it.
>>>>>>
>>>>>> In the case that it does not exist, is the usual parm94 method the
>>>>>> right
>>>>>> way
>>>>>> to go about making charges for this?
>>>>>>
>>>>> If I well understood your problem, you need an OME fragment
>>>>> representative
>>>>> of an _ester_ group:
>>>>>
>>>>> I do not think you can use the "OME" fragment available in the GLYCAM
>>>>> force field topology database (FFTopDB) (because the corresponding
>>>>> charge
>>>>> derivation has been carried out using the CHELPG algo. & this fragment
>>>>> has
>>>>> been designed to cap an _acetal_) and you cannot use the "NME" or "ACE"
>>>>> fragments available in the AMBER FFTopDB because they were designed to
>>>>> cap a
>>>>> peptide with two _peptide bonds_.
>>>>>
>>>>> However, you could follow a similar approach to that used to generate
>>>>> the
>>>>> NME or ACE chemical group or for the central fragment of an amino-acid:
>>>>> See for instance:
>>>>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#10
>>>>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#15
>>>>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#15
>>>>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
>>>>>
>>>>> You need to define an "intra-molecular charge constraint" (intra-mcc)
>>>>> set
>>>>> to zero for a chemical group you are going to remove, while you are
>>>>> going to
>>>>> keep the OME group. The definition of an "intra-mcc" is available in the
>>>>> section "-7th area-" in the resp manual.
>>>>> See http://q4md-forcefieldtools.org/RED/resp/ &
>>>>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#10
>>>>>
>>>>> The key point is to find the right molecule where you will apply this
>>>>> intra-mcc. You might start from an ester such as MeCO-OMe where you are
>>>>> going to define an intra-mcc set to zero for the acetyl group (MeCO);
>>>>> thus
>>>>> the total charge of your OME fragment will be an integer (zero) i.e.
>>>>> compatible with the Amber FFTopDB.
>>>>>
>>>>> If you decide to use the R.E.D. tools, this approach is straightforward:
>>>>> you only need to use the INTRA-MCC keyword in a P2N input file for the
>>>>> selected group of atoms involved in the constraint such as:
>>>>>
>>>>> REMARK INTRA-MCC 0.0 | 1 2 3 4 5 6 | Remove
>>>>> => set the intra-mcc to zero &
>>>>> remove the atoms numbers 1-6 (CH3CO group) from the FF library
>>>>> (mol2 file format)
>>>>> REMARK
>>>>> REMARK TITLE Ester...
>>>>> REMARK CHARGE-VALUE 0
>>>>> REMARK MULTIPLICITY-VALUE 1
>>>>> REMARK INTRA-MCC 0.0 | 1 2 3 4 5 6 | Remove
>>>>> REMARK
>>>>> ATOM 1 C1 EST 1 7.137 -3.656 0.065 C1
>>>>> ATOM 2 H11 EST 1 7.203 -4.172 -0.893 H11
>>>>> ATOM 3 H12 EST 1 7.873 -2.853 0.099 H12
>>>>> ATOM 4 H13 EST 1 7.333 -4.369 0.866 H13
>>>>> ATOM 5 C2 EST 1 5.738 -3.078 0.230 C2
>>>>> ATOM 6 O3 EST 1 5.589 -1.896 0.530 O3
>>>>> ATOM 7 O4 EST 1 4.754 -3.918 -0.044 O4
>>>>> ATOM 8 CT5 EST 1 3.441 -3.409 0.022 C5
>>>>> ATOM 9 H5 EST 1 3.250 -3.022 1.023 H51
>>>>> ATOM 10 H5 EST 1 2.731 -4.206 -0.202 H52
>>>>> ATOM 11 H5 EST 1 3.326 -2.605 -0.706 H53
>>>>> CONECT 1 2 3 4 5
>>>>> CONECT 2 1
>>>>> CONECT 3 1
>>>>> CONECT 4 1
>>>>> CONECT 5 1 6 7
>>>>> CONECT 6 5
>>>>> CONECT 7 5 8
>>>>> CONECT 8 7 9 10 11
>>>>> CONECT 9 8
>>>>> CONECT 10 8
>>>>> CONECT 11 8
>>>>> END
>>>>>
>>>>> You run R.E.D.-III.x & you directly get a Tripos mol2 file fragment for
>>>>> OME in the organic function you are interested in.
>>>>>
>>>>> Finally, the key is to look at the RRMS of the fit & check if the
>>>>> "intra-mcc" used does not break the fit; in this case, R.E.D.
>>>>> automatically
>>>>> fits with & without the intra-mcc; thus, you can easily compare both
>>>>> fits.
>>>>>
>>>>> Anyway, this MeCO-OMe model should provide you a reasonable starting
>>>>> point.
>>>>>
>>>>> regards, Francois
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Mar 02 2010 - 22:00:04 PST
