search

chanzuckerberg / cellxgene

An interactive explorer for single-cell transcriptomics data
https://chanzuckerberg.github.io/cellxgene/
MIT License
614 stars 116 forks source link

view results from pre-computed differential expression #855

Open neuromusic opened 5 years ago

neuromusic commented 5 years ago

As a scanpy user who has already computed differential expression via scanpy.tl.rank_genes_groups and saved the results to .uns['rank_genes_groups'], I want to view these results in cellxgene so that I can explore marker genes that differentiate labeled groups of cells.

See: https://scanpy.readthedocs.io/en/stable/api/scanpy.tl.rank_genes_groups.html

neuromusic commented 5 years ago

for context on how scanpy visualizes this data ...

import scanpy as sc
pbmc = sc.datasets.pbmc68k_reduced()

sc.tl.rank_genes_groups(
    pbmc,
    groupby='louvain'
)

sc.pl.rank_genes_groups(pbmc)

image

colinmegill commented 5 years ago

I love this and think we should do it. I'll noodle.

ambrosejcarr commented 3 years ago

The AnnData formalism for differential expression is more general than just scanpy algorithms. @vals said it would be helpful if cellxgene could bring in these kinds of data, since he can store his scVI differential expression results in this form.

Also relevant for chanzuckerberg/single-cell#15 -- this kind of data could be brought in via gene sets.