tx clustering demo preparation, presentation, feedback

[wbwakeman]

We should aim to do a demo of current transcriptomics clustering work and future plans to Zizhen's group the last week of April. This issue is to capture the effort required to prepare for this demo, to do the demo, and to address any feedback from the demo.

[MatthewAitken]

Per discussion, demo will include:

• No chunking, Tasic data only
• Data loading, normalization, means/variances, highly variant genes, and pca
• May include Louvain clustering if can be connected
• Notebook/package api and use demonstration
• Accuracy of outputs compared to scrattch.hicat R
• In development api including chunking and writing copies

Future demo will demonstrate:

• Chunking
• writing copies
• memory/speed performance metrics on large dataset
• Other currently unfinished modules