AMBER Archive (2009)

Subject: Re: [AMBER] non standard residue

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Tue Jul 14 2009 - 01:47:00 CDT


Dear Eduardo,

R.E.D. has detected a putative problem in the fitting step.
Consequently, the user is warned & the force field library is not
generated. This is indeed because you have all these 0.000000 values
in the "q(opt)" column.

I suspect a problem in the constraints you set up in the P2N input file.
Did you look at the examples of P2N file available from
http://q4md-forcefieldtools.org/Tutorial/ ?
For instance:
http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
For the Central fragment:
http://q4md-forcefieldtools.org/Tutorial/P2N/All-frag-Pept/Mol_red1.p2n
For the dipeptide itself:
http://q4md-forcefieldtools.org/Tutorial/P2N/All-frag-Pept/Mol_red6.p2n

  --

If this does not help:
This is difficult to understand the reason only from the RESP input
itself. May I ask you to send me your P2N input file & explain if you
want to derive the charges for the THR(-) dipeptide itself (i.e. the
whole molecule) or if you want to derive the charges for a THR(-)
fragment.

  --

Finally, if you decide to use R.E.D. @ Server
http://q4md-forcefieldtools.org/REDS/, you will be able to generate
the charges for the dipeptide molecule & the Central, N-terminal and
C-terminal residues in a single step. This is quite
simple/straighforward.

In your RESP input, you have a single conformation. You could also
imagine generating charges & FF library using various fitting
approaches based on different conformations as in the F-74 R.E.DD.B.
project.
See http://q4md-forcefieldtools.org/REDDB/projects/F-74/.
There you will find 18 molecules involved in charge derivation where
various conformations are used and considered as different molecules.

Finally, pay attention to the conformation(s) you are going to select
as your dipeptide presents a negative charge...

regards, Francois

> Dear Francois.
>
> I have been trying to use the RED program to obtain charges for a "new"
> residue (just a deprotonated THR),
> and so far RED fails at this stage:
> MEP(s) is/are being computed for molecule 1 ... [ OK ]
> See the file(s) "JOB2-gam_m1-1-(X).log"
>
> The RESP-A1 charges are being derived for molecule 1 ... [ FAILED ]
>
> Is this because I have no charges in the input?.
>
> What can I change to make the program work?.
> I am sorry if the question is way too general, but I am lost in here,
> any input would be appreciated.
>
> Eduardo Mendez
>
>
>
> ===================================================
> -----------------------------------------------
> Restrained ESP Fit 2.3 Amber 4.1
> -----------------------------------------------
> RESP-A1 project. RESP input generated by R.E.D.
> -----------------------------------------------
>
>
> inopt = 0 ioutopt = 1
> nmol = 1 iqopt = 1
> ihfree = 1 irstrnt = 1
> iunits = 0 qwt = 0.00050000
> wtmol(1) = 1.000000
> subtitle:
>
> THRn
> ich = -1 iuniq = 25
> 1 6 0
> 2 1 0
> 3 1 0
> 4 1 0
> 5 6 0
> 6 8 0
> 7 7 0
> 8 1 0
> 9 6 0
> 10 1 0
> 11 6 0
> 12 1 0
> 13 6 0
> 14 1 0
> 15 1 0
> 16 1 0
> 17 8 0
> 18 6 0
> 19 8 0
> 20 7 0
> 21 1 0
> 22 6 0
> 23 1 0
> 24 1 0
> 25 1 0
> IQOPT=1, all q0 values will be set to 0
>
> ----------------------------------------------------------------------------
> ATOM COORDINATES CHARGE
> X Y Z
> ----------------------------------------------------------------------------
> ----------------------------------------------------------------------------
>
> Charge on the molecule(ich) = -1
> Total number of atoms (iuniq) = 25
> Weight factor on initial charge restraints(qwt)= 0.50000E-03
>
> there are 1 charge constraints:
>
> 1 1
> 2 1
> 3 1
> 4 1
> 5 1
> 6 1
> 7 1
> 8 1
> 9 1
> 10 1
> 11 1
> 12 1
> 13 1
> 14 1
> 15 1
> 16 1
> 17 1
> 18 1
> 19 1
> 20 1
> 21 1
> 22 1
> 23 1
> 24 1
> 25 1
>
> Reading esp"s for molecule 1
> total number of atoms = 25
> total number of esp points = 0
>
> center X Y Z
> 1 0.6131043E+01 -0.2094224E+01 0.5506935E+00
> 2 0.7111837E+01 -0.2874522E+01 -0.1072238E+01
> 3 0.7437647E+01 -0.8464902E+00 0.1518306E+01
> 4 0.5581057E+01 -0.3616848E+01 0.1800480E+01
> 5 0.3900742E+01 -0.5536205E+00 -0.3923372E+00
> 6 0.4293678E+01 0.1303377E+01 -0.1709129E+01
> 7 0.1658620E+01 -0.1340652E+01 0.3691328E+00
> 8 0.1387931E+01 -0.3069192E+01 0.1219936E+01
> 9 -0.8105000E+00 -0.6920698E+00 -0.6093841E+00
> 10 -0.8161093E+00 -0.8860557E+00 -0.2656968E+01
> 11 -0.2626874E+01 -0.2837297E+01 0.6172498E+00
> 12 -0.3635130E+01 -0.1845102E+01 0.2153658E+01
> 13 -0.4634660E+01 -0.3570338E+01 -0.1350356E+01
> 14 -0.3720053E+01 -0.4503911E+01 -0.2939798E+01
> 15 -0.5928221E+01 -0.4924380E+01 -0.5036890E+00
> 16 -0.5708861E+01 -0.1961189E+01 -0.2038283E+01
> 17 -0.1263334E+01 -0.4794381E+01 0.1405457E+01
> 18 -0.1761214E+01 0.1969244E+01 -0.9934437E-01
> 19 -0.3950021E+01 0.2516949E+01 -0.4499880E+00
> 20 -0.7519792E-01 0.3713217E+01 0.7430303E+00
> 21 0.1724155E+01 0.3384996E+01 0.2885653E+00
> 22 -0.8462255E+00 0.6318471E+01 0.9267326E+00
> 23 -0.1337000E+01 0.7132906E+01 -0.8959781E+00
> 24 -0.2480558E+01 0.6481775E+01 0.2142703E+01
> 25 0.6905458E+00 0.7403727E+01 0.1739375E+01
> Initial ssvpot = 0.000
>
> Number of unique UNfrozen centers= 25
>
> Non-linear optimization requested.
> qchnge = 0.1109400188E-01
> qchnge = 0.2660590324E-08
> qchnge = 0.4710277376E-17
>
> Convergence in 2 iterations
>
> THRn
> Point Charges Before & After Optimization
>
> no. At.no. q(init) q(opt) ivary d(rstr)/dq
> 1 6 0.000000 0.000000 0 0.005000
> 2 1 0.000000 -0.076923 0 0.000000
> 3 1 0.000000 -0.076923 0 0.000000
> 4 1 0.000000 -0.076923 0 0.000000
> 5 6 0.000000 0.000000 0 0.005000
> 6 8 0.000000 0.000000 0 0.005000
> 7 7 0.000000 0.000000 0 0.005000
> 8 1 0.000000 -0.076923 0 0.000000
> 9 6 0.000000 0.000000 0 0.005000
> 10 1 0.000000 -0.076923 0 0.000000
> 11 6 0.000000 0.000000 0 0.005000
> 12 1 0.000000 -0.076923 0 0.000000
> 13 6 0.000000 0.000000 0 0.005000
> 14 1 0.000000 -0.076923 0 0.000000
> 15 1 0.000000 -0.076923 0 0.000000
> 16 1 0.000000 -0.076923 0 0.000000
> 17 8 0.000000 0.000000 0 0.005000
> 18 6 0.000000 0.000000 0 0.005000
> 19 8 0.000000 0.000000 0 0.005000
> 20 7 0.000000 0.000000 0 0.005000
> 21 1 0.000000 -0.076923 0 0.000000
> 22 6 0.000000 0.000000 0 0.005000
> 23 1 0.000000 -0.076923 0 0.000000
> 24 1 0.000000 -0.076923 0 0.000000
> 25 1 0.000000 -0.076923 0 0.000000
>
> Sum over the calculated charges: -1.000
>
> Statistics of the fitting:
> The initial sum of squares (ssvpot) 0.000
> The residual sum of squares (chipot) 0.000
> The std err of estimate (sqrt(chipot/N)) nan
> ESP relative RMS (SQRT(chipot/ssvpot)) nan
>
> Center of Mass (Angst.):
>
> X = 0.00000 Y = 0.00000 Z = 0.00000
>
> Dipole (Debye):
>
> X = -0.06007 Y = 0.02430 Z = -0.14783
>
> Dipole Moment (Debye)= 0.16141
>
> Quadrupole (Debye*Angst.):
>
> Qxx = -21.41502 QYY = -20.60831 QZZ = 42.02333
> Qxy = 55.28119 QXZ = 44.67650 QYZ = 45.26516
>
>
>
>
>
> FyD wrote:
>> Dear Neha,
>>
>>> I have a protein structure (protein.pdb) and it has a residue called ACA in
>>> chain C. I have prepared lib file (ACA.lib). Then I describe head and tail
>>> group of this non standard residue. I saved the file as ACA.lib. I m yet to
>>> calculate the charges.
>>>
>>> I try to load the whole protein after loading the lib file. The N-terminal
>>> of ACA residue connects well with the previous residue but the
>>> C-terminal of
>>> this residue misses out the connectivity with the N atom of the proceeding
>>> residue which is MET.
>>
>> Here you have to well define the head & the tail for your new
>> peptide fragment in your LEaP script. See for instance this example:
>> http://q4md-forcefieldtools.org/Tutorial/P2N/All-frag-Pept/script1.cmd
>>
>> For the charge derivation of amino-acid fragments & force field
>> library building, you can look at:
>> http://q4md-forcefieldtools.org/Tutorial/
>> 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#16
>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#17
>> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
>>
>>> Well, this residue is connected at the ends with standard amino acid and
>>> hence if I use the gaff forcefield, it complains about the atom
>>> type between
>>> standard and non standard residue. If I use amber atom type then its
>>> complains about the N-N bond type wherein no such bond exists in the
>>> peptide.
>>
>> Yes I saw examples of such N-N bond in the literature in peptide analogs:
>> you might look at the work of P. Legrel (Rennes, France)
>> You need to use the Amber99SB force field/force field atom types.
>> For this N-N bond, you will have to develop your own FF parameters.
>>
>> regards, Francois

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