Project

General

Profile

Feature #683

do_dssp could be improved

Added by Mark Abraham almost 9 years ago. Updated over 5 years ago.

Status:
Closed
Priority:
Low
Assignee:
Category:
analysis tools
Target version:
-
Difficulty:
uncategorized
Close

Description

do_dssp could report more data.

It doesn't report the average secondary structure content.

It doesn't write a matrix of secondary structure over time except as an .xpm file.

-ssdump dumps the secondary structure only for the first trajectory frame.

Associated revisions

Revision 07a84a86 (diff)
Added by Mark Abraham almost 9 years ago

Improved do_dssp -sc output

do_dssp -sc now computes the total for each secondary structure type.
This makes it easy to report the average secondary structure content
by type in a comment at the bottom of that file.

IssueID #683

Revision 5e8544f5 (diff)
Added by Mark Abraham almost 9 years ago

Improved do_dssp -ssdump output

do_dssp -sc now dumps rows of strings with the secondary structure
content for the whole trajectory. g_chi has a feature that reads in
this file, however g_chi is hard-coded to read only the first line,
so this change does not affect that feature.

IssueID #683

History

#1 Updated by Mark Abraham almost 9 years ago

  • Status changed from New to In Progress

#2 Updated by Justin Lemkul almost 9 years ago

Mark Abraham wrote:

do_dssp could report more data.

It doesn't report the average secondary structure content.

I like the new addition of this feature, but there's one problem. The do_dssp output factors in a "structure" term for the chain separator (symbol "="), such that it becomes part of the total percentage of secondary structure. If, for example, I run do_dssp on a dimeric protein that I've simulated, I get:

...


@ s0 legend "Structure" 
@ s1 legend "Coil" 
@ s2 legend "B-Sheet" 
@ s3 legend "B-Bridge" 
@ s4 legend "Bend" 
@ s5 legend "Turn" 
@ s6 legend "Chain_Separator" 
       0    57    15    52     2     8     3     1
# Totals    57    15    52     2     8     3     1
# SS %    0.70  0.19  0.64  0.02  0.10  0.04  0.01

The incorporation of this term undermines the utility of this feature, since I then have to go back and manually compensate for it anyway. I suppose the better solution is to remove the chain separator as a "structure" term, since it messes up the .xpm output (i.e. writing 81 residues instead of 80 in the above example), and doesn't matter for scount.xvg.

#3 Updated by Mark Abraham almost 9 years ago

I think this feature was present in do_dssp before I touched it. As far as I recall from the code, chain separators would have been added to totals in the second column before I added the functionality to sum columns.

I agree that as "chain separation" is not a secondary structure type it shouldn't go in the total, and that's not hard to implement, I don't want to make the change myself without further consultation. It could break existing analysis scripts.

I have added David and Carsten to the watchers, since they've implemented some of the features under discussion/modification.

#4 Updated by Justin Lemkul almost 9 years ago

Mark Abraham wrote:

I think this feature was present in do_dssp before I touched it. As far as I recall from the code, chain separators would have been added to totals in the second column before I added the functionality to sum columns.

Absolutely, I just wanted to make sure you (and anyone else that might jump on board) was aware of the chain separator thing, in case you hadn't tested it. Nice features, by the way; I'm glad to see them implemented.

I agree that as "chain separation" is not a secondary structure type it shouldn't go in the total, and that's not hard to implement, I don't want to make the change myself without further consultation. It could break existing analysis scripts.

I have added David and Carsten to the watchers, since they've implemented some of the features under discussion/modification.

#5 Updated by Rossen Apostolov over 5 years ago

  • Status changed from In Progress to Resolved

We can mark this as resolved. Please reopen if needed.

#6 Updated by Rossen Apostolov over 5 years ago

  • Status changed from Resolved to Closed

Also available in: Atom PDF