Jupytext looks like a good solution for version control of .ipynb notebooks, allowing clean version history in md or py and synchronising them with the rendered ipynb notebooks.

Use existing infrastructure from widget-ts-cookiecutter to set up back-end and front-end testing.

Start with Python I/O - see https://clustergrammer.readthedocs.io/matrix_format_io.html and https://github.com/ismms-himc/clustergrammer2/blob/master/examples/introduction_nb.ipynb

10X released a dataset of ~100K CD8 T Cells, we should make an example notebook working with this data. We need to be careful to not add too much data to this repo however. Make a branch for this.

Dataset link: https://www.10xgenomics.com/extra/#whats-new?utm_source=Twitter&utm_medium=&utm_content=&utm_campaign=2019-Social-Twitter

https://support.10xgenomics.com/single-cell-vdj/datasets/3.0

right now the tests fail if a dataset is changed or is empty or only has one row

• create a custom task file
• create fixed results to show

cornerstone has native support for layers and transparency and it would be useful to output segmentations and heatmaps and show multi-modal images like PET-CT

relevant JS is https://github.com/cornerstonejs/cornerstone/blob/803dd7335e16de3cc398a3d42eda2552dd47cb1b/example/layers/index.html#L95-L152

This should teach someone how to create an IG manually and also how to generate one in a more automated way using FHIR short hand.

Describe the bug

The labels for maps etc are now non-infomrative due to the use of {rlang}.

To Reproduce

Look at any plot using interactive = TRUE

Expected behavior

Labels should be informative + clear

Screenshots

<img width="1680" alt="Screenshot 2019-09-13 09 04 51" src="https://user-images.githubusercontent.com/20317649/64864742-c3d1d500-d605-11e9-90d3-7