I`ve been working with ccpNMR for about a month now, and am quite impressed with it`s stability and functionality, but there is a significant issue that has been bothering me for a few weeks now. (The workstation is a Dell machine running Ubuntu 20.04, with the sddm display manager)
I have assigned, for our protein, the backbone resonances and the HNCACB/HNCACBCO, HNCO/HNCACO, HNCOCA/HNCA combinations, and am working through assignment by protein sequence assignment (I am not in front of the workstation as of yet, and will update this post if I have added an incorrect name/label/window).
The program will display certain probabilities of the amino acid sequences and residue identities, and will show the possible identities of the I, I+1 and I-1 residues properly (if I select one of the matches, but do not form an I+1/I-1 link. When I actually assign a link, some of these possibilities disappear - as well as the possible sequence identity window, with no indication as to why. Removing the link does not `bring back` the previous possible identities for the sequence or the amino acid residue that the program suggested before I made the link.
Any suggestions would be greatly appreciated on fixing this problem/trying to pinpoint it. I will be happy to provide whatever details necessary, although taking screenshots is difficult at the moment, as the previous app I have tried ignored the ccpNMR windows (and took a picture of the desktop behind them)!
Hi William,
Thanks for pointing this out. We`ll take a look at it.
Vicky
Hi William,
sorry to have taken a while to get back to you. The strange thing is that I have a dim and distant memory of something like this happening to me several years ago (well before I joined the CCPN team). I never reported it at the time, as it didn`t particularly bother me - I tended not to use the predictions, as it annoyed me that you could always only see a small window of residues at a time. Matching strings of linked residues to a sequence was one of the few parts of my work where I tended to use pen and paper! But bizarrely, I can`t now seem to recreate the issue.
So I can`t really offer much in the way of help. And given that we are trying to concentrate our development resources on V3 rather than V2, I fear we probably wouldn`t rate this issue high enough to devote much time to it. Sorry.
In terms of amino acid type predictions, I tend to work with this in my head:
Gly - CA in the 40s
Ala - `high` CA and CA `above` 20 (I work visually here - high/above on the screen, but smaller in terms of numbers!)
Thr/Ser - `low` CB and usually you you can tell the difference between them because only Thr goes `below` 70
I/V/P - all have very `low` CAs (60s)
N/D/L - CA and CB are `close together` (CAs near 50, CBs in the low 40s)
F/Y - `low` CAs (low 50s-60s) and `low` CBs (40s)
D/E/R/K/P/M/W/H - `high` CBs (in the 30s)
C - depends on the oxidation state
Once grouping residues things into these categories, you can usually match things to the seqeuence quite nicely.
All the best,
Vicky