Error in Interactive when going through the tutorial

[jefferyUstc]

Hi, nice work of spatialdata to facilitate the integrtive analysis of multi-modal spatial omics.

I'm checking out the tutorial provided in the documentation and encountered an Error below:

Tutorial: Use landmark annotations to align multiple -omics layers, https://spatialdata.scverse.org/en/latest/tutorials/notebooks/notebooks/examples/alignment_using_landmarks.html#adding-landmark-annotations

After read in the visium and xenium data, I run the Interactive function to visualize the data:

Interactive([visium_sdata, xenium_sdata], points=False, shapes=False)

• Error 1 occurs :

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[12], line 1
----> 1 Interactive([visium_sdata, xenium_sdata], points=False, shapes=False)

TypeError: Interactive.__init__() got an unexpected keyword argument 'points'

• Error 2 occurs:

 Interactive([visium_sdata, xenium_sdata])

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[4], line 1
----> 1 Interactive([visium_sdata, xenium_sdata])  # , points=False, shapes=False

File ~/miniconda3/envs/SpatialData/lib/python3.10/site-packages/napari_spatialdata/_interactive.py:220, in Interactive.__init__(self, sdata)
    218 self._viewer = napari.Viewer()
    219 self._sdata = sdata
--> 220 self._sdata_widget = SdataWidget(self._viewer, sdata)
    221 self._list_widget = self._viewer.window.add_dock_widget(
    222     self._sdata_widget, name="SpatialData", area="left", menu=self._viewer.window.window_menu
    223 )

File ~/miniconda3/envs/SpatialData/lib/python3.10/site-packages/napari_spatialdata/_interactive.py:56, in SdataWidget.__init__(self, viewer, sdata)
     52 self._viewer = viewer
     54 self.setLayout(QVBoxLayout())
---> 56 self.coordinate_system_widget = CoordinateSystemWidget(self._sdata)
     57 self.elements_widget = ElementWidget(self._sdata)
     59 self.layout().addWidget(QLabel("Coordinate System:"))

File ~/miniconda3/envs/SpatialData/lib/python3.10/site-packages/napari_spatialdata/_interactive.py:42, in CoordinateSystemWidget.__init__(self, sdata)
     38 super().__init__()
     40 self._sdata = sdata
---> 42 self.addItems(self._sdata.coordinate_systems)

AttributeError: 'list' object has no attribute 'coordinate_systems'

More information:

(SpatialData) ➜  ~ pip list | grep spatialdata             
napari-spatialdata            0.2.4
spatialdata                   0.0.10
spatialdata-io                0.0.4
spatialdata-plot              0.0.2

[timtreis]

@rahulbshrestha do you have an idea why that is happening?

[rahulbshrestha]

@timtreis This notebook was based off the "spatialdata" branch in napari-spatialdata. The above error is because it used the main branch.

We don't support multiple spatialdata objects in the main branch of the Interactive class right now (scverse/napari-spatialdata#112), which would fix this issue.

[jefferyUstc]

@timtreis This notebook was based off the "spatialdata" branch in napari-spatialdata. The above error is because it used the main branch.

We don't support multiple spatialdata objects in the main branch of the Interactive class right now (scverse/napari-spatialdata#112), which would fix this issue.

Thank you, I see, hope to see the latest updated documentation soon. once I read the manu from biorxiv, I can't wait to use the framework.

[jefferyUstc]

BTW, when i run other tutorials, I also got some bugs. I guess it's also raised by the conflicting versions in the releasing ones (pypi) and the ones used in the documentation?

For example:

• technology_visium
  https://spatialdata.scverse.org/en/latest/tutorials/notebooks/notebooks/examples/technology_visium.html

just load the data and render the image:

import spatialdata_plot

visium_sdata.pl.render_images().pl.render_shapes().pl.show("ST8059049")

It outputs:

• Spatial query
  https://spatialdata.scverse.org/en/latest/tutorials/notebooks/notebooks/examples/spatial_query.html

the last step when overlay the selected region:

f, ax = plt.subplots(figsize=(7, 7))
sdata_ST8059048.pl.render_shapes(color="library", palette="Pastel1").pl.show(ax=ax)
cropped_sdata.pl.render_shapes(color="Sox2").pl.show(ax=ax, colorbar=False)

# show the spatial query bounding box
rect = patches.Rectangle((8000, 12000), 4000, 4000, linewidth=5, edgecolor="k", facecolor="none")
ax.add_patch(rect)
ax.invert_yaxis()

It outputs:

---

FYI:

anndata 0.9.1
dask 2023.5.1
dask-image 2023.3.0
h5py 3.8.0
imageio 2.31.1
matplotlib 3.7.1
matplotlib-inline 0.1.6
matplotlib-scalebar 0.8.1
napari-matplotlib 0.2.1
napari-spatialdata 0.2.4
networkx 3.1
Pillow 9.5.0
scanpy 1.9.3
scikit-image 0.21.0
shapely 2.0.1
spatial_image 0.3.0
spatialdata 0.0.10
spatialdata-io 0.0.4
spatialdata-plot 0.0.2

if any further info needed, donnot hesitate to contact me.

[timtreis]

Hey @jefferyUstc, thank you for pointing these bugs out! Yeah, we're currently in a bug-squashing phase which hasn't translated to a new pypi release yet. I'll test the aggregation bug and potentially tag you in the issue over at spatialdata_plot if that's where the bug comes from. For the static plotting in particular, we found and fixed quite a few bugs in the last few weeks. So you might want to just use the github main branch for the most stable version so far :)

[LucaMarconato]

@jefferyUstc thanks for reporting. As mentioned before the problems with napari were caused by the fact that the spatialdata branch was used for visualizing the data with napari and not the main branch. And as @timtreis said, most of the spatialdata-plot bugs should be fixed by now.

The following issues keep track of the bugfix on napari:

• Test reproducibility notebooks napari-spatialdata#121
• Test documentation notebooks napari-spatialdata#122
• Make a function to open napari with every layer shown napari-spatialdata#123

There should be only one remaining bug with spatialdata-plot, tracked here:

• get_extent produces wrong output after spatial query spatialdata-plot#100

Since now the problems are tracked elsewhere, I will close the current issue. But if you find new problems that are not tracked by the open issues above, please feel free to reopen this issue.