Complete the ranges selection in CING to What If, CING etc.

[GoogleCodeExporter]

What steps will reproduce the problem?
1. E.g. running a CASD prediction like:
http://nmr.cmbi.ru.nl/CASD-NMR-CING/data/GR/CGR26AFrankfurt/CGR26AFrankfurt.cing
The ranges are:
56-60,63-93,98-111,115-168

What is the expected output? What do you see instead?
I saw that Procheck is working fine. I noted that What If  isn't capable of 
taking a region. The 
solution could be to chop and feed it only the region of interest but it will 
score worse for those 
checks that expect atoms to be present; e.g. inside/outside and hydrogen bond 
checks.

For the percentage of residue ROG in the summary the ranges are taken well into 
account:
http://nmr.cmbi.ru.nl/CASD-NMR-
CING/data/GR/CGR26AFrankfurt/CGR26AFrankfurt.cing/CGR26AFrankfurt/HTML/summary.h
tml

Original issue reported on code.google.com by jurge...@gmail.com on 10 May 2010 at 8:31

[GoogleCodeExporter]

Needed for overall numbers for sure.

Original comment by jurge...@gmail.com on 11 Nov 2010 at 10:21

• Added labels: Priority-Critical
• Removed labels: Priority-Medium

[GoogleCodeExporter]

I thought I had reported this elsewhere but for the record now:

We need to change the string specifying ranges from using a dash '-' to using 
colon ':'.
That code is in ranges2list and asci2list

Original comment by jurge...@gmail.com on 16 Nov 2010 at 1:09

[GoogleCodeExporter]

This is important for almost 400 entries that have negative residue numbers.

SELECT distinct e.pdb_id FROM "nrgcing"."cingresidue" r,  "nrgcing"."cingentry" 
e where
r.entry_id = e.entry_id AND
r.number < 0
order by e.pdb_id

Original comment by jurge...@gmail.com on 16 Nov 2010 at 4:10

[GoogleCodeExporter]

Committed to r876 for Geerten's review. 

Original comment by jurge...@gmail.com on 16 Nov 2010 at 4:16

• Changed state: Started

[GoogleCodeExporter]

Geerten made me aware of the need of including chain ids in here as well. I 
will code for one of the possibilites he suggested:

[chainId1.]resNum1:resNum2

eg:A.20:40 

In the absence of a chain id the selection should expand to all matching 
residues in potentially multiple chains.

Original comment by jurge...@gmail.com on 17 Nov 2010 at 10:42

[GoogleCodeExporter]

Suggestion is to switch to Yasara's format for ranges with negative numbers:

-5--1 is valid
-5 - -1 is equivalent (spacing between numbers is not important).
-5 -- 1 is invalid becaus the final dash needs to be attached to the digit (1).

For chains this would become:
A.-5--1 is valid
A.-5-A.-1 is invalid

- Usually we will leave out the optional spaces ok? So we prefer -5--1 over -5 
- -1.

- We prefer to write: A.1-10,B5-15 instead of 1-15 even if this would 
technically be the same.

Original comment by jurge...@gmail.com on 23 Nov 2010 at 2:13

[GoogleCodeExporter]

Implemented new range selection in r879.

Original comment by jurge...@gmail.com on 24 Nov 2010 at 10:53

[GoogleCodeExporter]

For now we'll use my old method of selecting ranges of well-defined segments 
described in: Doreleijers et al. Quality assessment of NMR structures: a 
statistical survey. J Mol Biol (1998) vol. 281 (1) pp. 149-164

In short: backbone cv window averaged over 3 residues below 0.2.

Original comment by jurge...@gmail.com on 24 Nov 2010 at 11:04

[GoogleCodeExporter]

-We prefer to write: A.1-10,B5-15 instead of 1-15 even if this would 
technically be the same.
 needs to be:
-We prefer to write: A.1-10,B.5-15 instead of 1-15 even if this would 
technically be the same.

Original comment by jurge...@gmail.com on 2 Dec 2010 at 10:04

[GoogleCodeExporter]

Implemented and tested well now.

Original comment by jurge...@gmail.com on 13 Dec 2010 at 9:39

• Changed state: Fixed