> 1. according to you, I changed the atomtype CM for C6 to C4 and checked
> with rdparm command.I found that the chi parameters are applying. but the
> problem is that there are to atoms C6 and C5, both have atom type CM. Then
> what I did is that I have also changed the atom type CM to C4 for other
> bonds, angles and torsions parameters corresponding to atom name C6 only
> leaving the two torsion parameter for chi (O4'-C1'-N1-C6 AND C2'-C1'-N1-C6
> ACCORDING TO YOUR PARM99CHI PAPER) and also leaving the atom name C5 as it
> is. Now what happened is that the resulting prmtop file is taking only the
> chi parameters for your frcmod.parmCHI file and all other parameters from
> 2SU.frcmod file. As I found from SantaLucia paper they have optimized (by
> QM) the whole modified residue and as result of that all the parameter for
> 2SU is different from U in normal parm99.dat file.
Do not change anything except the atom type of atom name C6. It might mess
up your prmtop. Atom types are defined to describe particular properties.
Atom named C5 and C6 (in uridine) have the same properties except the chi
torsion. C5 does not have such a chi term.
You have to read some basics about MD simulations and force fields.
If you talk about the following SantaLucia paper
Aduri, R., Psciuk, B. T., Saro, P., Taniga, H., Schlegel, H. B.,
SantaLucia, J., Jr. (2007) "AMBER Force Field Parameters for the
Naturally Occurring Modified Nucleosides in RNA", J. Chem. Theory and
Comput., 3, 1464 -1475.
the main work of this paper is about resp charge calculations. The general
amber force field (gaff) was then used to assign the missing parameters
(bond/angle/dihedral). They did not do any torsional fitting for the
modified residues. For resp calculation, you still need to optimize the
structure using a QM program and then calculate the molecular
electrostatic potentials.
Your best approach would be to combine santalucia's .prepi and frcmod with
revised chi torsions. In order to make sure that revised chi's are used
for 2SU, you need to change the atom type of atom named C6. There might be
some missing parameters that you might need to include in your frcmod file
(such as the ones you were asking in your 2nd question). Use the amber99
parameters for only those missing terms. See parm99.dat file.
> 2. If I only change the atomtype CM to C4 for atomname C6 in the 2SU.prepin
> file, tleap ask for two torsions (C6-N1-C2-N3 AND C6-N1-C2-S2) and one
> angle parameter (C6-N1-C2) which I have to provide from 2SU.frcmod file.
> But chi parameter is taking from your frcmod.parmCHI file and all other
> from parm99.dat file. It is like mixing up all the parameters and may be
> logical.
There will be some missing terms. Make analogy with the previously
defineds parameters. For example, C6-N1-C2 angle is CM-N*-C (according to
amber99 force field). With the revised chi version, this will be C4-N*-C.
Look at parm99.dat and find what the parameters are for CM-N*-C and use
those values for C4-N*-C.
>
> 3. I found that that without the revised chi parameter the normal
> nucleosides (with parm99.dat) as well as the modified nucleosodes (with
> Santa Lucia parameters) are not reproducing the experimental conformational
> distribution. You found that revised chi parameter is reproducing the
> experimental observations. So my logic is that to combine only the revised
> chi parameters of yours with the modified nucleoside parameters (developed
> by Santa Lucia group). If my logic is correct then issue-1 is more suitable
> than issue-2. What is your suggestion ???
See my comments above. Without the chi revision, you will probably see a
domination of syn conformations.
Best,
Ilyas,
k
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Dec 17 2012 - 13:00:04 PST
