Viewer: Structure-Sequence colouring

I have multiple protein structures superimposed and did some colouring of residues corresponding to each other. Next, I display sequences for those superimposed proteins and colour these sequences by the colouring I did in the 3D window. Yet, further manipulation of the sequences/alignment, for instance introducing insertions by entering whitespace or deleting them is highly laggy when one works with tens of structures (>50) at once... Apparently there's much real-time communication between the 3D window and sequence window which generates unnecessary computational load. Do you have any ideas on how to un-lag this process?

Answers 6
TY
TY 2021-06-28

Hi Kamil, it’s difficult to say what the exact cause is, as there may be many factors involved. It’s possible that this could be a memory issue.
I run Discovery Studio 2021 on 64bit Windows 10, 32 GB RAM.  I don’t have your structures or the exact steps you are doing, but I tested using the CDK2 Structures and Sequences.dsv file (50 structures) from the Samples folder and didn’t notice any appreciable lag with adding or deleting spaces in the sequence window.
You might check how much RAM you have on your computer and consider adding more memory if possible.  Are you using a Nvidia graphics card or Integrated graphics?

KS
KS 2021-06-29

Thank you for your answer. I'm using several machines, with 16, 24 and 32 GB of RAM. I use GeForce or quadro. CPUs i7 or Xeon. The hardware itself is quite decent. 

Did you test that with an option to colour alignment background by structure (in order to connect the colouring of the structure to alignment)?

TY
TY 2021-06-30

Hi Kamil, yes I have tested using the color alignment option – from the Sequence window > Display Style > Color Scheme (Background) set to 3D Structure:Protein Display. I didn’t observe the slowdown.
As a test, I would suggest that you use the CDK2 Structures and Sequences.dsv file from the Samples folder and describe the exact sequence of steps you performed which resulted in the slow response. I will try to reproduce the issue from my side.

KS
KS 2021-07-05

I read the sample project and do:

right lick on 3d window -> show all
right click on the alignment window -> display style -> color scheme -> background -> 3D structure: Protein Display

and now introducing insertions / removing insertions to sequence alignment gets laggy.
When I show only one structure on the screen, sequence manipulation gets faster, but the colouring of hidden sequences disappears (and I want to have the colouring present).

TY
TY 2021-07-06

Hi Kamil. Thanks for the update, sorry but I still cannot reproduce the problem either with add/move insertions to sequence alignment or coloring by Protein Display in the sequence window.
I have tested:  Show All in 3D window.  Sequence > Secondary Structure > Visibility, choose Kabsch and Sander method.  Then I made some color changes in the 3D window: for the first long helix (Leu101 to Ser120) I pick the color purple, lime green, pink for proteins 1AQ1, 1B38, 1E1V respectively.
In the sequence window, Display Style > color scheme > background > 3D Structure: Protein Display.  As expected, I see the three new colors displayed for this helix section in the sequence window.  Now I turn off the display for 1AQ1, 1B38, 1E1V in the 3D window, the coloring in the sequence window remains.
Please see my screenshot below showing the 3D window and the linked sequence window.  See that I turned off the first three protein display in the 3D window, as a result the red X displayed to the left of the sequence names in the sequence window.  The color changes I’ve made do remain.

comment imagecomment image
JS
JS 2021-07-12

Hi Kamil, I took a look at this issue and I do see the problem you see. The thing is when you hide the molecule in the Molecule Window it is then no longer linked to the sequence and so it can't retain the coloring based on the molecule coloring (that coloring is meant to be dynamic meaning that any change in coloring in the Molecule Window will automatically change the color in the Sequence Window). I think I have come up with a workaround. This attached Discovery Script can be used to color the sequence by the linked molecule's color but in this case the coloring is not dynamic. I think it could reduce the lag you are seeing as well. However, you should note that if you change the coloring in the Molecule Window you will need to rerun the script in order of the Sequence Window coloring to be updated. 
To use this script, from the DS Client go to the Files Explorer, right-click on the unzipped  ColorSequenceByProteinDisplayAlternative.pl file and select Open With | Script Window then launch the script using the Run command from the Scripting toolbar. It assumes you already have a Molecule Window and Sequence Window loaded with the linked objects. The script will update the coloring of the Sequence Window.

comment imagecomment image