CHARMM Development Project
Hi, I am simulating a peptide having alternating L and D amino-acid in lipid bilayer (DMPC). So I use the input generator for protein/membrane complex in CHARMM GUI and I run standard mdp files in gromacs.(force field that I have opt for is CHARMM36)
>
> My aim for simulation is to observe the dihedral angle of each residue in ramachandran plot.
>
> Error that I am getting-
> My peptide is 14 residue peptide and everything run flawlessly but after extracting the data for dihedral angle and plotting it I can see one D amino acid residue is in second quadrant of ramachandran plot. It's a disallowed region for D amino-acid residue right?
>
> I am attaching a drive link here in which there is a screenshot of DALA-6 ramachandran plot and there you can see some dihedral angle come in second quadrant and some in 3rd which is not allowed region for D aminoacid.

and in LD14_2_updated folder that i have attached there are all files including initial structure, mdp files that i have generated by charmm gui.
link-
https://drive.google.com/folderview?id=12D-ikoZYjbNN09XxpBZfvkbLYa8dX49U

please have a look and do comment if something seems wrong.

>
I have done the whole simulation in gromacs
> A quick reply would be appreciated
>
If you think it's a CHARMM-GUI problem, you should use the Contact page on their site.

If you think it's a force field problem, you may have to show that with some simple tests that demonstrate the issue more clearly than you have done here.

We cannot help you with Gromacs problems, or problems with the force field conversion.
Ok,
One thing I wanted know can this kind of switch happen during simulation?

This is not a gromacs problem nor forcefield as I have erformed simulation for smaller peptide like 2residue ,6 residue ,10 residue and 12 residue there I got good ramification. That's why I am bewildered.

Is there any way to rebuild the Hydrogen or position the Hydrogen according to D chirality ?
No, such switches do not normally happen. Did you specify in any way that you have D-amino acids in you system? This is necessary - even though the standard terms in the force field are symmetric wrt L- and D-amino acids, the CMAP cross terms are different and so you need to use toppar_all36_prot_c36_d_aminoacids.str and the residue names in it; this is in CHARMM, I have no idea about CHARMM-GUI or GROMACS.
Yes I do have specified while generating input in charmm-gui.

Most imperative question is that ; why only one D amino acid residue switch while all other 5 D amino acid residue acquire their position in allowed region .

I have attached the drive link please have a look in LD14_2_updated folder in it where there is sub folder named TOPPAR consisting of forcefield.itp and PROA.itp which has all the paramater and CMAP.

Can you please have a look?

I am a tyro in this field that's why facing problem.


Thank you in advance
You should probably consult GROMACS support.
Sorry to say but they have told me to ask here in CHARMM forum. In that way i am not getting help 😒.
Chiral inversion can happen, and often does in various enhanced sampling methods, so it is something to check for.

The best way to do that is to compute the improper dihedral, defined as the angle between two planes determined by 3 atoms: for amino acids, one plane is defined by N,CA,C and the other by N,CB,C and the angle should be ca. +120 or -120 depending on the chirality.

Besides enhanced sampling, a lack of careful model building can also cause chiral inversions, for example if there are large forces from bad contacts where some atoms are too close to each other.
Ok thank you I will look into it.

One thing I observe is that conformation may not be switching from D to L but after I plot the phi psi angle in ramachandran plot I see a some data goes in 2nd quadrant for D amino acid and some data in usual allowed region for D residue.

How can we explain this ?
How to compute the improper dihedral mentioned by you ?
I'm not exactly sure what you mean by the quadrant numbers, it's not a definition I'm familiar with.

I've only used CHARMM to compute the improper dihedral, but it's just straightforward geometry for two planes defined by 3 atoms, with 2 common atoms (N,C) defining the axis.
You must be acquainted about Ramachandran plot.

There are 4 quadrant right, so I am saying that some frames of simulation of D RESIDUE comes in second quadrant which is disallowed region according to GN ramachandran for D amino acid
I am well aware of the plot and quadrants, but not the precise definition of which quadrant is numbered 1, 2, etc.; do you have a reference (URL or citation) for the numbering scheme?
https://images.app.goo.gl/kTswwMCbzfbWi3gN6

https://images.app.goo.gl/w42HjjYV9Ajuiyyr6


I have attached a URL please have a look. First one is allowed region for L and D residue

Second one state the quadrant numbering

Quadrant nomenclature or numbering of ramachandran plot is same as Cartesian coordinate
© CHARMM forums