-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Lingering Issues for First Public Release #86
Comments
From a performance perspective, only the components whose start and/or end bins fall within the selected bin range should be rendered. New components on the right are rendered when their start bin is included in the range, and the components on the left disappear when their end bin are excluded. This lead to an up/down movement when the shift lead to load new rendered LinkArrows: it is not beautifull, but the performance are improving. When only the rearrangements are shown, it is not clear how many columns to shift for each bin unit changes. Maybe the components should be entirely visualized or not but the feedback is not good (you can't know how many time to click to hide a component because it is not visible how long it is). Perhaps in this display mode, the unit must become the component and not the bin? |
"only the rearrangements" is a bug. It's not being calculated correctly in leftXstart. Component width calculated in the JSON "x" is basically useless in this mode, so ignore component.columnX. Instead, the padding for a whole chunk is JSON.chunk.component_count and JSON.chunk.link_count. For components inside of a chunk, the X will have to be manually calculated with a cumulative sum. I was considering removing the "Show Only Rearrangements" checkbox for now until this gets fixed. It's better that every feature available is bug-free. Unless this bug is resolved quickly, we can just ensure it's not accessible today. |
Now, when only rearrangements are shown, the X start should be correct. I am still not sure how to shift in this visualization mode. When only rearrangements are shown, the component width does not correspond to the number of bin, so I can't shift it 1 bin at time. For now, it is avoided any shift and each component appears/disappears entirely when its EndBin is/isn't in the visualized bin range. |
@josiahseaman, the implementation can certainly be improved, and will be done in the future, but I think the functional parts of the issue are done and ready for tests. |
I've now confirmed CS nucleotide slicing has one extra nucleotide. It just doesn't show up in small_test because the fasta file ends. If you add more, then you'll see it. |
Great job guys. I am experiencing a much better browser feeling now! I took some time and played with the tool. Here are some comments from my side:
Questions from my side:
I know it looks like a lot, but its mostly minor stuff. Anyhow, you really improved this a lot since the hackathon!!! These are just comments and suggestions. I understand if we would not be able to solve all of them. |
…dth 1; added pangenome last bin position for the user; code cleaning; #86
For me I have some comments:
|
Lingering issues for Public Release #86
…rsor icon changes when it is on the navigation bar #86
Wow! What an exciting feature! I think you should recruit a second person
to test it, maybe Dymtro? Then we could merge it to master and deploy.
I've been thinking about provenance coloring a lot lately. Being able to
add in paths for gene annotations would be a game changer for our release...
…On Tue, Jun 23, 2020 at 8:19 AM Andrea Guarracino ***@***.***> wrote:
Now the navigation bar is clickable to go to a specific region with a
click.
[image: image]
<https://user-images.githubusercontent.com/62253982/85415120-3a283e00-b56d-11ea-842b-1f5ac2550f9d.png>
—
You are receiving this because you modified the open/close state.
Reply to this email directly, view it on GitHub
<#86 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AARG2FEE3YKDNV46K7JAE2LRYC2X5ANCNFSM4NEH35OQ>
.
|
@dimatr, I wait for you with open arms, destroy all my changes. @josiahseaman, I want to point out that in my last commits I assumed the use of newer odgi versions, where there is a slightly change in odgi bin when the bin width is 1, so the dataset should be update before merge the last changes to master and deploy. The provenance coloring will be the next step (by the way, will it this the format for metadata?). |
Lingering issues:
From Dmytro's list
From Simon's list
Column Width
leads to ugly arrows:Use vertical compression
leaves some grey space at the bottom which does not correspond to any path, I think.npm start
gives me a compilation warning:Browserslist: caniuse-lite is outdated. Please run next command 'npm update'
<-- Updated the packages' json filesThe text was updated successfully, but these errors were encountered: