HI Anselm,
I checked into this and found that you're right about the atom types in Tyr
(same for Phe). For these aromatic side chains we actually did not replace
the atom type with 2C as we did for the others. Changing the others was
sufficient to isolate them. The parameters in the library files are correct
and were fit to these amino acids. What is incorrect is the SI Tables S1
and S2 that list 2C instead of CT. We will prepare an updated SI and check
with the journal about an erratum so that the parameters are described
properly in the paper.
thanks for the careful analysis and excellent documentation of the issue!
Carlos
On Tue, Jun 8, 2021 at 5:57 AM Dr. Anselm Horn <anselm.horn.fau.de> wrote:
> Dear all,
>
> while trying to create ff14SB parameters for a tyrosine derivative, I
> came across something in the parameter files I do not understand:
>
> In the ff14SB publication (Maier et al. 2015) referenced in the Amber21
> manual, it is reported that new atom types are introduced and used when
> needed; for TYR, Supplement Tables S1 and S2 list chi torsional angles
> that comprise two new atom types, namely CX and 2C.
>
> While I find the atom type CX for the TYR carbon alpha atom in the
> ff14SB parameter files, I cannot find 2C for the carbon beta, despite
> its description in the paper.
>
> Since I intend to stay as close as possible to the original ff14SB TYR
> parameters for my TYR-derivative, I'm not sure which the actual ff14SB
> parameters are for TYR - maybe I missed something completely.
> Thus, any clarification is greatly appreciated!
>
> I used AmberTools version 21.03, and Amber version 20.11.
> Starting from leaprc.protein.ff14SB, I inspected the parameter files
> loaded there, parm10.dat and frcmod.ff14SB; the later holds the
> definition of the 2C atom type and the corresponding parameters.
> The actual residue description resides in amino12.lib, which lacks a 2C
> atom type for TYR (using CT for carbon beta instead), but contains 2C
> atom types for other standard residues (ASN, ASP, CYS, GLN, GLU, ILE,
> LEU, MET, SER). BTW, the situation seems similar for PHE.
> For carbon beta of other amino acids with ring systems (HIS and TRP),
> however, the force field library files and the paper description do
> agree: CT is used for carbon beta.
>
> I'm looking forward to hints from the experts!
>
> Best regards,
>
> Anselm
>
>
>
>
>
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 08 2021 - 11:30:02 PDT
