plotting data horizontal plots with shared axis

[rbf22]

• [x ] I have checked that this issue has not already been reported.
• [ x] I have confirmed this bug exists on the latest version of scanpy.
• (optional) I have confirmed this bug exists on the master branch of scanpy.

---

The problem here is that the violin plots are horizontal rather than vertical and that they share the same x-axis scale. Plotting two sets of numbers 10s vs 10,000s the 10s data are not observable.

This looks like a change in the way that seaborn wants violin plots called, based solely on the error below.

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Minimal code sample (that we can copy&paste without having any data)

sc.pl.violin(adata, ['n_genes_by_counts', 'total_counts', 'pct_counts_house'],
             jitter=0.4, multi_panel=True)

/Users/banana/opt/miniconda3/lib/python3.8/site-packages/anndata/_core/anndata.py:1192: FutureWarning:

is_categorical is deprecated and will be removed in a future version.  Use is_categorical_dtype instead

/Users/banana/opt/miniconda3/lib/python3.8/site-packages/seaborn/_core.py:1303: UserWarning:

Vertical orientation ignored with only `x` specified.

Versions

WARNING: If you miss a compact list, please try print_header!

anndata 0.7.4
scanpy 1.6.0
sinfo 0.3.1

PIL 8.0.1
anndata 0.7.4
appnope 0.1.0
attr 20.2.0
backcall 0.2.0
cffi 1.14.0
cycler 0.10.0
cython_runtime NA
dateutil 2.8.1
decorator 4.4.2
get_version 2.1
h5py 2.10.0
idna 2.9
ipykernel 5.3.4
ipython_genutils 0.2.0
ipywidgets 7.5.1
jedi 0.17.2
joblib 0.17.0
jsonschema 3.2.0
kiwisolver 1.2.0
legacy_api_wrap 1.2
llvmlite 0.34.0
matplotlib 3.3.2
mpl_toolkits NA
natsort 7.0.1
nbformat 5.0.8
numba 0.51.2
numexpr 2.7.1
numpy 1.19.1
packaging 20.4
pandas 1.1.2
parso 0.7.1
pexpect 4.8.0
pickleshare 0.7.5
pkg_resources NA
plotly 4.12.0
prompt_toolkit 3.0.8
ptyprocess 0.6.0
pvectorc NA
pygments 2.7.1
pyparsing 2.4.7
pyrsistent NA
pytz 2020.1
retrying NA
scanpy 1.6.0
scipy 1.5.2
seaborn 0.11.0
setuptools_scm NA
sinfo 0.3.1
six 1.15.0
sklearn 0.23.2
statsmodels 0.12.1
storemagic NA
tables 3.6.1
tornado 6.0.4
traitlets 5.0.5
wcwidth 0.2.5
zmq 19.0.2

IPython 7.18.1
jupyter_client 6.1.7
jupyter_core 4.6.3
notebook 6.1.4

Python 3.8.3 (default, May 19 2020, 13:54:14) [Clang 10.0.0 ]
macOS-10.15.7-x86_64-i386-64bit
4 logical CPU cores, i386

Session information updated at 2020-11-05 17:47

[giovp]

Hi @rbf22 ,
what's the issue here?

[rbf22]

Hi @rbf22 ,
what's the issue here?

I’ve updated the issue. Thanks!

[giovp]

mmh a preoducible example means something like this:

adata = sc.datasets.pbmc68k_reduced()
adata.var["mt"] = adata.var_names.str.startswith(
    "MT-"
)  # annotate the group of mitochondrial genes as 'mt'
sc.pp.calculate_qc_metrics(
    adata, qc_vars=["mt"], percent_top=None, log1p=False, inplace=True
)
sc.pl.violin(
    adata,
    [
        "n_genes_by_counts",
        "total_counts",
    ],
    jitter=0.4,
    multi_panel=True,
)

and this works.

what's 'pct_counts_house' that you are calling?

[rbf22]

Pct_counts_house could be replaced by mt in your example. I’ll try you example and post the result.

[rbf22]

This does not fail, but it does produce the warnings:

/Users/banana/opt/miniconda3/lib/python3.8/site-packages/seaborn/_core.py:1303: UserWarning: Vertical orientation ignored with only x specified.

Here is the plot that shows that individual plots are horizontal compared to the previous versions that were vertical. Also you can see that the plots share the same x axis, which should not be the case as a default for scanpy I suspect. To see how scanpy has historically plotted this data check out: https://scanpy-tutorials.readthedocs.io/en/latest/pbmc3k.html

This was generated with seaborn version 0.11.0 (sept 2020) https://seaborn.pydata.org/whatsnew.html

[rbf22]

Try this for a fix,

import seaborn as sns
sns.set_theme(style="whitegrid")
tips = sns.load_dataset("tips")
ax = sns.violinplot(x=tips["total_bill"])

ax = sns.violinplot(y=tips["total_bill"])

So to have the vertical plots the syntax has changed.

[rbf22]

The line that creates this in scanpy looks like:

    g = g.map(
        sns.violinplot, y, orient='vertical', scale=scale, order=keys, **kwds,
    )

[WeilerP]

@rbf22, if I am not mistaken, this issue has been fixed in #1422.

[rbf22]

Yes, this looks like the same issue, happy to close this for now. Is matplotlib more stable for plotting?

[Dan-121]

sorry, i miss the point, How can we fix the problem, I met the same problem

[giovp]

try updating scanpy

[Dan-121]

Thank you,I will have a try  发自我的iPhone

…

------------------ Original ------------------ From: giovp <notifications@github.com&gt; Date: Mon,Jan 4,2021 10:45 PM To: theislab/scanpy <scanpy@noreply.github.com&gt; Cc: Dan <735393152@qq.com&gt;, Comment <comment@noreply.github.com&gt; Subject: Re: [theislab/scanpy] plotting data horizontal plots with shared axis (#1486) try updating scanpy — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.