Stop pointing at scanpy-tutorials

[ivirshup]

Moved links for plotting tutorials to point at scanpy docs, since these tutorials are hosted there now.

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[ivirshup]

Oof, is this going to have to re-run the notebook since I updated a line?

[grst]

yes, that's expected... and now everything falls apart, which is also somewhat expected since nobody executed the notebooks in half a year.

[ivirshup]

I can probably make a release of scanpy that exposes the sc.external.pp.scrublet but just redirects to sc.pp.scrublet, then revert changes to the notebook? We probably should be redirecting for that anyways.

[grst]

yeah that could work as a quick fix. But at some point we'll have to take a look at the notebooks either way.

[flying-sheep]

Ah, so all 6 are executing because the cache no longer exists?

The actual error we run into seems to be https://github.com/scverse/muon/pull/136 anyway

[flying-sheep]

@grst seems like without a cache entry, this fails. What shall we do?

[grst]

5/6 notebooks have passed, not so bad.

The actual error is

 ---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In [40], line 1
----> 1 dc.run_mlm(mat=adata, net=markers, weight=None, source="cell_type", target="genesymbol", verbose=True, use_raw=False)

File /usr/share/miniconda3/envs/tutorials/lib/python3.10/site-packages/decoupler/method_mlm.py:116, in run_mlm(mat, net, source, target, weight, batch_size, min_n, verbose, use_raw)
    114 # Transform net
    115 net = rename_net(net, source=source, target=target, weight=weight)
--> 116 net = filt_min_n(c, net, min_n=min_n)
    117 sources, targets, net = get_net_mat(net)
    119 # Match arrays

File /usr/share/miniconda3/envs/tutorials/lib/python3.10/site-packages/decoupler/pre.py:149, in filt_min_n(c, net, min_n)
    146 net = net[msk]
    148 if net.shape[0] == 0:
--> 149     raise ValueError("""No sources with more than min_n={0} targets. Make sure mat and net have shared target features or
    150     reduce the number assigned to min_n""".format(min_n))
    152 return net

ValueError: No sources with more than min_n=5 targets. Make sure mat and net have shared target features or
        reduce the number assigned to min_n
ValueError: No sources with more than min_n=5 targets. Make sure mat and net have shared target features or
        reduce the number assigned to min_n

in the basic tutorial. I might be able to take a look tomorrow, but can't guarantee.

[ivirshup]

@grst, I think we may want to revert this and rethink a little. The cluster labellings in the "preprocessing and clustering" tutorial are now incorrect.

I think it is highly unlikely that bottom group contains:

• B cells
• Proerythroblasts
• And CD4+ T cells

I think it's all going to be B cells

Here are CellTypist's predictions, and enrichment scores using Pangalao markers. Both show that grouping to be B-cells