Support #112
MaxCMO
| Status: | New | ||
|---|---|---|---|
| Priority: | Normal | ||
| Assignee: | - | ||
| Category: | - | ||
| Target version: | - | ||
| Resolution: |
Description
Hi I recently ran an ProCKSI experiment and have been slowly digesting the results.
I came across a result in MaxCMO that is difficult to understand.
I'm comparing a family of proteins with a similar fold. One particular pdb 5ue8 chain A and B
were compared using MaxCMO. This is the same protein and structure just a different chain. MaxCMO does not align these proteins correctly and states there is 5% sequence identity between these proteins. 5ue8 A and B are identical proteins with several residues in one chain not represented in the other chain.
Is this a flaw in how MaxCMO aligns 5ue8-A against 5ue8-B?
I would very much appreciate your advice on how i should interpret these results.
here is the link to the page " http://www.procksi.net/data/5580955b2e2df751d1f90557e8e84d11/MaxCMO/0_1/0_1.al "
Could you also extend the time limit on the expiration of these results as it is taking me a long time to fully understand the results .
Declan
History
#1
Updated by Paweł Widera almost 6 years ago
The methods used by ProCKSI compare protein structures, not sequences. Max-CMO in particular, maximises the number of overlaps between contacts (with default settings: pairs of C-alpha atoms not farther from each other than 0.75nm). The alignment is structural, not sequence-driven. This is why you should look at the overlap value or number of alignments in the results, rather than the sequence identity. This is what the distance matrices in the Results Manager show:
http://www.procksi.net/data/5580955b2e2df751d1f90557e8e84d11/MaxCMO/results_MaxCMO.html
You can download all the results to your computer before the experiment expires. We are unable to store it long term, sorry.