RNA velocity analysis component

[LuckyMD]

Components needed for RNA velocity analysis and visualization inclusion into open pipelines:

1. QC:
   function:
   scv.utils.show_proportions(adata)

• just to see if data has sufficient unspliced counts

2. Preprocessing and fit:
   functions:
   scv.pp.filter_and_normalize(adata, min_counts=20, min_counts_u=10)
   scv.pp.moments(adata)
   scv.tl.recover_dynamics(adata)

3. QC2: Gene likelihood check:

• this is manual so far...
• some function on: adata.var[‘fit_likelihood’]
• assessment (distribution plot, table for certain cutoffs)
• potentially look at phase portraits of potential velocity genes
• set adata.var['velocity_genes’] as mask on adata.var[‘fit_likelihood’]

4. Velocity calculation step 2:
   function:
   scv.tl.velocity(adata, mode=‘dynamical’)

• other modes possible, but this is probably best

5. Velo visualization calculation:
   functions:
   scv.tl.velocity_embedding(adata, basis=‘umap’)

• other bases possible, like diffmap, phate, etc)

6. Velocity visualization plotting:
   function(s):
   scv.pl.velocity_embedding(adata, basis=‘umap’, color=, arrow_size=2, arrow_length=5, alpha=0.2, size=20)

• also other plots like velocity_embedding_grid or velocity_embedding_stream
• should combine into 1 visualization component ideally