Closed JobLeonard closed 7 years ago
So my thoughts are to first sit down and redesign the UI (well, first first to fix a few small remaining bugs and upload that version to the server). We've created something almost useful by now, but the number of bells and whistles will soon overwhelm the user if we don't take a moment to step back and get the sidepanels and such cleaned up.
I think after/as part of that I'll tackle the "paste genes" option, because that is relatively straightforward and directly useful.
Then I think of working on the lasso-selection, because that is actually quite a big issue that breaks down into a lot of sub-issues; along the way of implementing it quite a few other useful features (proper axes & interactive canvases, for example) will have to be built as a side-effect. So it's a good "milestone" to work towards.
Sounds good @slinnarsson?
Yes, sounds good!
I liked this D3 “polybrush” selection tool: http://bl.ocks.org/christophermanning/5020993
/Sten
Sten Linnarsson, PhD Professor of Molecular Systems Biology Karolinska Institutet Unit of Molecular Neurobiology Department of Medical Biochemistry and Biophysics Scheeles väg 1, 171 77 Stockholm, Sweden +46 8 52 48 75 77 (office) +46 70 399 32 06 (mobile)
On 22 Nov 2016, at 13:21, Job van der Zwan notifications@github.com<mailto:notifications@github.com> wrote:
So my thoughts are to first sit down and redesign the UI (well, first first to fix a few small remaining bugs and upload that version to the server). We've created something almost useful by now, but the number of bells and whistles will soon overwhelm the user if we don't take a moment to step back and get the sidepanels and such cleaned up.
I think after/as part of that I'll tackle the "paste genes" option, because that is relatively straightforward and directly useful.
Then I think of working on the lasso-selection, because that is actually quite a big issue that breaks down into a lot of sub-issues; along the way of implementing it quite a few other useful features (proper axes & interactive canvases, for example) will have to be built as a side-effect. So it's a good "milestone" to work towards.
Sounds good @slinnarssonhttps://github.com/slinnarsson?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/linnarsson-lab/Loom/issues/71#issuecomment-262227245, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AKKag0MM3lqSHBaP4Z4XzHo4Z28LJ8A2ks5rAt5XgaJpZM4K5VRS.
Oh, nice! Haven't looked at the code yet, but that looks like an elegant way to do it interface-wise! Thanks for the suggestion!
My interpretation of these notes, not in order of priority:
Website in general
password access
look for gene enrichment -> requires lasso select
"What would Patrick want from this site?"
More interactivity than Python plots
UI/UX redesign to prevent context switches even more
Sparklines
Scatterplot
paste list of genes, like sparklines
add edges to SFDP
manual scaling size of dots relative to guess
low expression -> small dots, high expression -> larger dots
matrix of colors x genes
compare genes side-by-side
lasso-selection of clusters
Heatmap
multiple attributes visible at once
"different color ramps please!" - Gioelle
zooming in weird aspects ratios?
allow for sorting rows/columns by attribute(s) before generating a heatmap (or, for that matter, loomfile)
Command Line