Clustergrammer-GL is a JavaScript library that produces interactive heatmaps using the WebGL library regl. Clustergrammer-GL's primary use is as dependency of Clustergrammer2, which is a Jupyter Widget data analysis and visualization library built to analyze single cell high-dimensional biological datasets. Our goals are:
- improve the interactive visualizations produced by Clustergrammer-GL and the Jupyter Widget Clustergrammer2
- improve the front-end data analysis capabilities of Clustergrammer-GL (e.g. front-end re-clustering)
- integrate Clustergrammer-GL into Clustergrammer-Web to enable visualizations of larger datasets on the web application
Last updated: July 26th, 2019
Clustergrammer-GL is primarily being developed to facilitate the analysis of single cell biological datasets and has certain default behaviors (e.g. row gene information lookup on mouseover, enrichment analysis).
- export genes to Enrichr: #18
We are setting up generalizable callback functions that will enable two-way communication to the Python kernel using the Jupyter Widget infrastructure.
- allow callback functions https://github.com/ismms-himc/clustergrammer-gl/issues/new?assignees=&labels=&template=feature_request.md&title=
Clustergrammer-GL is primarily being built as a dependency of Clustergrammer2, a single-cell data analysis toolkit. However, we are developing Clustergrammer-GL to be generalizable. We want Clustergrammer-GL to be useful for exploring datasets outside of biology (e.g. financial), embeddable into other contexts (e.g. custom web applications), and independent of a Python back-end (e.g. run clustering on JavaScript]). Issues associated with these goals are:
- category bleedthrough
- dendrogram slicing of linkage matrix
- camera update
- slow scrolling speed
- row/column search
- improve text cutoff
- dendrogram interactions
- tooltip overflow
- callback functions on common interactions
- initial ordering button
- category updating for Enrichrgram
- brush select zoom
Please see Clustergrammer's full documentation to see how Clustergrammer-GL fits into the larger project, which includes: