Anything to do about plt.subplot about scirpy

[UboCA]

I just realized scirpy has used plt.subplot to draw abundance bar plot.
So reusing plt.subplot may not work.

But how could I merge several plot objects from scirpy?

[UboCA]

Other observation while I using ir.pl.group_abundance is that legend will be resorted by single pair cell number.

ir.pl.group_abundance(adata, groupby='chain_pairing', target_col='seurat_clusters', normalize=True, sort 
= 'count')

You can see seurat_clusters {0...8} are reordered by chain_pairing value.
But what if one want to keep original catgeories? For this reason, to bring an order parametre could make it more friendly.

[grst]

Hi @UboCA,

thanks for reporting this issue. Actually, I'm a bit surprised that this hasn't surfaced earlier.
With the fix in #379, you should be able to do

adata = ir.datasets.wu2020_3k()
fig, (ax1, ax2) = plt.subplots(1,2, figsize=(12, 4), gridspec_kw={"wspace": 1})
ir.pl.group_abundance(adata, groupby='cluster_orig', target_col='patient',ax=ax1)
ir.pl.group_abundance(adata, groupby='cluster_orig', target_col='source',ax=ax2)

Could you give it a try? You can install the development version using

pip install git+https://github.com/scverse/scirpy@master

---

Regarding your second suggestion, there's a long-outstanding issue for that: #186
It should be an easy fix and I'm happy to take a PR. Otherwise I'll look into this once the mega PR #356 is complete.

[UboCA]

Got it, many thanks.

Best