- Posts: 5
Visualize virtual atoms in Martini3 Trp
- fralav
-
Topic Author
- Offline
- Fresh Boarder
Less
More
2 years 10 months ago #8998
by fralav
Visualize virtual atoms in Martini3 Trp was created by fralav
Hi everybody, I'm trying to simulate a small peptide in water containing the residue Trp by using topology and structure file generated with martinize according to martini 3 ff. Minimization, equilibration and md production run without errors, but when I convert my trajectory with trjconv for analysis I get the following error repeated many times:
There were 2 inconsistent shifts. Check your topology
By looking at the final trajectory the dummy atom correspondent to the virtual site of Trp side chain remains in the same position when the rest of the molecule moves freely in the box (picture at www.dropbox.com/s/9v3v348t1lakwh2/virtual-site.png?dl=0 )
I don't understand if this problem is really related to my topology or it's just a "structure visualisation problem" related to the way gromacs treats virtual sites (i'm using gromacs v 4.6.3). Even if it seems reasonable that this atom should not move because it has mass=0 I would like to know how to fix this.
I would also ask why martinize didn 't generate dihedrals parameters S1 BB BB S1 for my protein (except for the improper dihedrals related to the planar aromatic side chains).
thank you very much in advance for the support!
There were 2 inconsistent shifts. Check your topology
By looking at the final trajectory the dummy atom correspondent to the virtual site of Trp side chain remains in the same position when the rest of the molecule moves freely in the box (picture at www.dropbox.com/s/9v3v348t1lakwh2/virtual-site.png?dl=0 )
I don't understand if this problem is really related to my topology or it's just a "structure visualisation problem" related to the way gromacs treats virtual sites (i'm using gromacs v 4.6.3). Even if it seems reasonable that this atom should not move because it has mass=0 I would like to know how to fix this.
I would also ask why martinize didn 't generate dihedrals parameters S1 BB BB S1 for my protein (except for the improper dihedrals related to the planar aromatic side chains).
thank you very much in advance for the support!
Please Log in or Create an account to join the conversation.
- peterkroon
-
- Offline
- Gold Boarder
Less
More
- Posts: 210
2 years 10 months ago #9001
by peterkroon
Replied by peterkroon on topic Visualize virtual atoms in Martini3 Trp
> i'm using gromacs v 4.6.3
Is there a reason you're using such a prehistoric version? It contains bugs that were fixed in later versions.
Which S1-BB-BB-S1 dihedral(s) are you missing specifically? What command line did you use to generate the itp files?
Is there a reason you're using such a prehistoric version? It contains bugs that were fixed in later versions.
Which S1-BB-BB-S1 dihedral(s) are you missing specifically? What command line did you use to generate the itp files?
Please Log in or Create an account to join the conversation.
- fralav
-
Topic Author
- Offline
- Fresh Boarder
Less
More
- Posts: 5
2 years 10 months ago #9002
by fralav
Replied by fralav on topic Visualize virtual atoms in Martini3 Trp
Than you for your answer and for your help.
The dummy atom problem seems fixed by using later gromacs version.
Regarding the peptide topology, I used the following command:
martinize2 -f peptide.pdb -x peptide-CG.pdb -o peptide.top -ss CCCSSCBTTBTTC
My peptidic sequence contain several sequencial residues described by BB beat and one or more SC beats therefore I would expected that my itp file would contain also several SC1-BB-BB-SC1 dihedral parameters, in addition to BB-BB-BB-BB dihedrals, but I only have improper dihedrals for Trp and Tyr side chains. I noticed that also lipid's bonded interactions are only described by bonds and angles in the relatives itp files. Are then this dihedrals described implicitely?
Thank you very much again for your support
The dummy atom problem seems fixed by using later gromacs version.
Regarding the peptide topology, I used the following command:
martinize2 -f peptide.pdb -x peptide-CG.pdb -o peptide.top -ss CCCSSCBTTBTTC
My peptidic sequence contain several sequencial residues described by BB beat and one or more SC beats therefore I would expected that my itp file would contain also several SC1-BB-BB-SC1 dihedral parameters, in addition to BB-BB-BB-BB dihedrals, but I only have improper dihedrals for Trp and Tyr side chains. I noticed that also lipid's bonded interactions are only described by bonds and angles in the relatives itp files. Are then this dihedrals described implicitely?
Thank you very much again for your support
Please Log in or Create an account to join the conversation.
- peterkroon
-
- Offline
- Gold Boarder
Less
More
- Posts: 210
2 years 10 months ago #9008
by peterkroon
Replied by peterkroon on topic Visualize virtual atoms in Martini3 Trp
Martini uses relatively few dihedrals (only where strictly necessary), so there's a good chance those are absent on purpose. But please double check with the relevant publications! It could also be a bug in martinize2...
Please Log in or Create an account to join the conversation.
Time to create page: 0.095 seconds