Project

General

Profile

Bug #398

Interpretation of MainChain is incorrect for ffamber

Added by Justin Lemkul over 9 years ago. Updated over 9 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Erik Lindahl
Category:
mdrun
Target version:
Affected version - extra info:
Affected version:
Difficulty:
uncategorized
Close

Description

I was running DSSP analysis on some trajectories simulated with AMBER03. The screen output accurately reported that 40 residues were found in the "MainChain" group, which is the entire length of my peptide. However, ss.xpm contained data for only 39 residues. After comparing this all-atom simulation with one done under OPLS-AA, I found that the C-terminal oxygens, OC1 and OC2, were not being considered in the analysis.

I patched several source files and was able to fix the output. I'm still a C novice, so I hope these fixes are suitable. Also missing were "OT" atom names in the definition of "MainChain" in src/gmxlib/index.c; applicable fixes that solved my issue were:

src/gmxlib/index.c (starting at line 285 in the git version of the code):

static const char *mc[]      = { "N","CA","C","O","O1","O2","OT","OC1","OC2","OXT" };
static const char *mcb[] = { "N","CA","CB","C","O","O1","O2","OT","OC1","OC2","OXT" };
static const char *mch[] = { "N","CA","C","O","O1","O2","OT","OC1","OC2","OXT",
"H1","H2","H3","H" };

src/tools/do_dssp.c (starting at line 194 in check_oo):

for(i=0; (i<atoms->nr); i++) {
if (strcmp(,"OXT")==0)
*atoms->atomname[i]=OOO;
else if (strcmp(
(atoms->atomname[i]),"O1")==0)
atoms->atomname[i]=OOO;
else if (strcmp(
(atoms->atomname[i]), "OC1")==0)
*atoms->atomname[i]=OOO;
}

The OC1, OC2, and OT atom names are also missing in share/top/defselection.dat, but I think this file is unrelated to the DSSP fix. The corrected lines in that file would be:

"MainChain" protein and name N CA C O O1 O2 OT OC1 OC2 OXT;
"MainChain+Cb" protein and name N CA CB C O O1 O2 OT OC1 OC2 OXT;
"MainChain+H" protein and name N CA C O O1 O2 OT OC1 OC2 OXT H1 H2 H3 H "[0-9]?H";

History

#1 Updated by Rossen Apostolov over 9 years ago

It's fixed, thanks Justin!

Also available in: Atom PDF