Enable user to define >1 cohort for subsequent visualization

[enemeth19]

[amaraim22]

"is there a way to look at expression of a specific gene within two cohorts that have different drivers side by side? For example: I am interested in PRKCA expression in EGFR mutant as well as KRAS mutant Lung Adenocarcinoma (LUAD). I would like to look at PRKCA expression in ΔEGFR and PRKCA expression in ΔKRAS patients, both within one heat map or violin plot. When selecting the cohorts of interest, the software appears to only consider patients that have BOTH mutations, ie. EGFR + KRAS (the final heatmap only has data from 1 patient). I know that I can alternatively download csv files and graph the raw data in prism for each condition separately, but I was just curious if I could avoid the step when looking at multiple drivers (10+)."

[enemeth19]

Consider using Materialize Carousel in UI for multiple cohorts (see example)

[enemeth19]

Anyone have any thoughts on something like this as just a first step in the UI? The idea is that if the user wanted to define a second cohort, they would scroll only the section of the page that is involved in "Customize your Cohort(s)" horizontally. Nothing else on the page would move/change. The same select boxes would appear on the second page. (This was made using Materialize's Carousel)

[ZhenghaoZhu]

We could try out the tutorial library suggested in #307 to improve the UI of this. @enemeth19 Where is the code for the carousel? I'll try to play around with the library and make a small sample tutorial.

[enemeth19]

That sounds great! The code is in the multipleCohorts branch