Documentation for spica_top.json / .top to .pdb ?

[alphataubio]

(1) Where can i find documentation for what CHL1, ASM, LSM, BSM, BENA, ... are in spica_top.json ? Some of them can be seen at https://www.spica-ff.org/forcefield.html but i cant find anything else at the spica-ff.org website or the journal publications even in supplementary materials.

(2) Also in lipid tutorial, it says to use a PDB file for lipids in Packmol but only .top files are available (except for the only one 1DOPC.pdb). I dont see a top2pdb.py script in spica-tools. Should i write one to contribute ?

Initial configuration of a DOPC bilayer using packmol
Packmol is a tool to place molecules in a simulation box with a desired geometry such as lipid bilayer and vesicle. A molecular structure of a single molecule should be given as an input. In this example, we make a DOPC bilayer in aqueous solution, for which we prepare the PDB files for a single DOPC molecule and a single CG water particle. A sample input script for packmol is attached here, which generate a membrane of 128 DOPC with 1300 water particles.

[yskmiyazaki]

@alphataubio
Thank you for your trying spica-tools and the lipid tutorial.

(1) We are sorry for not mentioning what those residue names correspond to.
"CHL1" corresponds to an normal cholesterol molecule whose resname comes from an atomistic force field, CHARMM.
"*SM"s are sphingomyelin molecules and the prefix of the resnames shows the length of the tail chain connecting to the amide bond of the molecules.
For "BENA", it might be a fatty acid used before SPICA was called SDK but not available with the current SPICA. So I should have removed it from the json.
I will arrange the json file add some information so that users can understand what the molecules are, and try to remove the unavailable molecules from it.

(2) Unfortunately, we do not have any tools to generate PDB files from .top files.
As you know for .top files, all information in the files are in the json file so it is easy to extract one molecule information for generating them.
For now, CG PDBs in SPICA simulations are always prepared through "cg_spica map2cg", namely, AA-to-CG mapping.
It would be useful if we have some PDB libraries for single molecules (typical lipids) and a tool to get their PDB files on a console (e.g. "cg_spica top2pdb").
As I am almost the only developer of this spica-tools and cannot devote enough time to its development, such contribution in support of SPICA are welcome .

[alphataubio]

im trying to create a spica 400nm spherical lipid bilayer system using PACKMOL with cholesterol 60%, SM 20%, cardiolipin 16%, PC 3% and PG 1%.

was it you or someone else that created the lipid tutorial ? do you know how the file 1DOPC.pdb was created ?

https://www.spica-ff.org/tutorial_lipid/packmol/1DOPC.pdb

[yskmiyazaki]

@alphataubio
S. Seo, an author of the SPICA FF paper, initially created the tutorial.
I guess he prepared an atomic DOPC lipid bilayer system with CHARMM-GUI first.
Then he extracted a single DOPC lipid from the bilayer PDB file and mapped it to the SPICA coarse-grain level.
When generating such single CG lipid pdb for PACKMOL, we basically use CHARMM-GUI for obtaining AA lipid PDB, having a good molecular shape to form bilayer structure, to be mapped to CG.
Or use Jmol, Discovery Studio if CHARMM-GUI does not have the lipid we want to simulate.

[xiaoyu2260]

I had two problems:

1. When I use the “ENM” command, I get a fig1 error. How can I fix this?
2. I found a problem when I use Small Molecule Protein (PDB ID:2k8x), there seems to be a problem with the topology file generated by the “ENM” command. The generated topology file seems to generate key information that should not be there. see fig2

Has anyone else encountered this problem? What is the solution?

[yskmiyazaki]

These problems have nothing with the documentation for spica_top.json.
So please open another issue to separate the problems after checking my first answers.

1. Could you share your input files used in the ENM command with us? Finding out the problem from the command-line without any inputs is a bit difficult for me.
2. Please apply the ENM command for a single protein structure in PDB format prepared by separating each of the peptides (chain A and B) from the original PDB file. The last paragraph at https://www.spica-ff.org/tutorial_protein2.html can be related to what you find.