(GSoC) Generic Assay Categorical/Binary data implementation Frontend

[Djokovic0311]

Summary

• Student Name: Jiahang Li
• Mentors: Gaofei Zhao, Ryan Fu
• Issue: Generic Assay Categorical/Binary data
• Backend PR: Implementation, Enhancement

Changes proposed in this pull request:

1. Add tabs to display three types of Generic Assay Data profiles grouped by their generic assay data type, applying a drop down list to select each profile to render in the tab.
   

2. Implement Generic Assay Categorical Data Display as below:
   

• On the left, there is a table displaying data enrichments fetched from backend endpoint for categorical data. In this table, we can see each entity's pvalue and qvalue. Other columns are set as default values such as Sample and Chi-suqared Test.
• When clicking on a certain row of the table, there will be a plot container including four choices on the right, showing this selected entity's distribution in multiple ways (. Several checkboxes are here for plot settings such as including NA data or not.

3. Implement Generic Assay Binary Data Display as below:

• Similar to Categorical Data, there is also a table containing categorical data enrichment. Otherwise, each row includes other feature columns such as enriched in which group and Co-occurrence Pattern (only for two-group analysis).
• For two group analysis, initially there are two plots showing Log2 ratio and altered frequency in 2 groups.
• Generally, there is a bar plot to display top number of entities in the table under multiple judgement, such as the ones with highest pvalue. A button named "Select Entities" allows people to choose how many columns will appear in the plot.

4. Implement Result view Comparison Tab as below
   
   The content is the same as above.

Could you please have a look at this pr? @dippindots @fuzhaoyuan Any suggestions or modifications are welcome!

[dippindots]

@Djokovic0311 It's looking good to me, just have a few minor feedback. It's also good to see if we can update these comments from the product team:

• Consistent naming for y-axis labels (change altered frequency -> frequency)
• In category tab always show first plot (first row selected by default)

We might want to add feature flag for our feature too, this feature flag will check for url parameters if we want to show a specific feature. It's a way we can merge pr first and then let the product team test the feature in an easy way.
See the example here: https://github.com/cBioPortal/cbioportal-frontend/blob/master/src/pages/groupComparison/GroupComparisonStore.ts#L667-L669

• Add feature flag for genericAssayBinaryTabShowable and genericAssayCategoricalTabShowable

[dippindots]

@Djokovic0311 Congrats! I just merged the fronted part of the project. Although it's still behind the feature flag, now the product team can test this easily. I will remove the feature flag after the product team says it's ok to release. Thanks for your hard work this summer!!

[Djokovic0311]

@dippindots Thank you again for your guidance! Grateful for the opportunity to work on this during GSoC and for having you as a mentor. Looking forward to more collaborations in the future!