sc.pl.matrixplot index error

Since I update to scanpy==1.4.5.1 I am getting multiple plotting errors. sc.pl.rank_genes_groups() and sc.pl.violin() are still working fine but I am getting errors in the rank_genes functions like sc.pl.rank_genes_groups_violin() and sc.pl.tracksplot()

##still working fine
sc.tl.rank_genes_groups(adata, 'louvain', method='wilcoxon')
sc.pl.rank_genes_groups(adata, n_genes=25, sharey=False)
result = adata.uns['rank_genes_groups']
groups = result['names'].dtype.names

pd.DataFrame(
    {group + '_' + key[:1]: result[key][group]
    for group in groups for key in ['names', 'pvals']}).head(5)
##gives error
sc.pl.rank_genes_groups_violin(adata, groups='2', n_genes=8)

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
~/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py in _ensure_valid_index(self, value)
   3169             try:
-> 3170                 value = Series(value)
   3171             except:

~/anaconda3/lib/python3.7/site-packages/pandas/core/series.py in __init__(self, data, index, dtype, name, copy, fastpath)
    273                 data = _sanitize_array(data, index, dtype, copy,
--> 274                                        raise_cast_failure=True)
    275 

~/anaconda3/lib/python3.7/site-packages/pandas/core/series.py in _sanitize_array(data, index, dtype, copy, raise_cast_failure)
   4160         if isinstance(data, np.ndarray):
-> 4161             raise Exception('Data must be 1-dimensional')
   4162         else:

Exception: Data must be 1-dimensional

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
<ipython-input-23-ccdbf8b7836c> in <module>
----> 1 sc.pl.rank_genes_groups_violin(adata, groups='2', n_genes=8) ## 200316 error fix later, also when I run the entire script from the start

~/anaconda3/lib/python3.7/site-packages/scanpy/plotting/_tools/__init__.py in rank_genes_groups_violin(adata, groups, n_genes, gene_names, gene_symbols, use_raw, key, split, scale, strip, jitter, size, ax, show, save)
    727             if issparse(X_col): X_col = X_col.toarray().flatten()
    728             new_gene_names.append(g)
--> 729             df[g] = X_col
    730         df['hue'] = adata.obs[groups_key].astype(str).values
    731         if reference == 'rest':

~/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py in __setitem__(self, key, value)
   3114         else:
   3115             # set column
-> 3116             self._set_item(key, value)
   3117 
   3118     def _setitem_slice(self, key, value):

~/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py in _set_item(self, key, value)
   3188         """
   3189 
-> 3190         self._ensure_valid_index(value)
   3191         value = self._sanitize_column(key, value)
   3192         NDFrame._set_item(self, key, value)

~/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py in _ensure_valid_index(self, value)
   3170                 value = Series(value)
   3171             except:
-> 3172                 raise ValueError('Cannot set a frame with no defined index '
   3173                                  'and a value that cannot be converted to a '
   3174                                  'Series')

ValueError: Cannot set a frame with no defined index and a value that cannot be converted to a Series

Versions:

scanpy==1.4.5.1 anndata==0.7.1 umap==0.3.10 numpy==1.18.1 scipy==1.4.1 pandas==0.23.0 scikit-learn==0.21.3 statsmodels==0.11.0 python-igraph==0.7.1 louvain==0.6.1 scvelo==0.1.25 scanpy==1.4.5.1 anndata==0.7.1 loompy==3.0.6 numpy==1.18.1 scipy==1.4.1 matplotlib==3.1.3 sklearn==0.21.3 pandas==0.23.0

[ivirshup]

@Maarten-vd-Sande, it looks like the issue here is a bug with handling dense data. With your particular example here, you should be fine to work around this by using a sparse matrix.

from scipy import sparse
All.X = sparse.csr_matrix(All.X)

[Maarten-vd-Sande]

@ivirshup I never replied, sorry. Your solution indeed fixed it for us. Thanks!