LouisFaure
commented
4 months ago Can you try with a previous version of adjustText? such as 0.7.3
Closed quantumdot closed 4 months ago
LouisFaure
commented
4 months ago Can you try with a previous version of adjustText? such as 0.7.3
quantumdot
commented
4 months ago Downgraded to adjusttext==0.8, and now the exception I reported above is avoided, and the plots now look as they are supposed to. However, I now encounter a different exception:
IndexError Traceback (most recent call last)
Cell In[24], line 1
----> 1 g=scf.pl.trends(adata, plot_emb=False, highlight_features=[], ordering="max",
2 fig_heigth=6, pseudo_cmap=flat_huslmap, ord_thre=.95, return_genes=True)
3 scf.pl.matrix(adata, features=g, nbins=20,
4 cmap="RdBu_r", annot_top=False,
5 colorbar_title="max\nnormalized\nactivity")
File c:\Users\username\AppData\Local\miniconda3\envs\scFates\lib\site-packages\scFates\plot\features.py:452, in trends(adata, features, cluster, highlight_features, n_features, root_milestone, milestones, module, branch, annot, title, feature_cmap, pseudo_cmap, plot_emb, plot_heatmap, wspace, show_segs, basis, heatmap_space, offset_names, fontsize, style, ordering, ord_thre, figsize, axemb, show, output_mean, save, return_genes, **kwargs)
447 axheatmap.add_patch(patch)
448 axheatmap.hlines(
449 fitted_sorted.shape[0], 0, fitted_sorted.shape[1], color="k", clip_on=True
450 )
--> 452 adjust_text(
453 texts,
454 ax=axheatmap,
455 add_objects=[patch],
456 va="center",
457 ha="left",
458 autoalign=False,
459 expand_text=(1.05, 1.2),
460 lim=5000,
461 only_move={"text": "y", "objects": "x"},
462 precision=0.1,
463 expand_points=(1.2, 1.05),
464 )
466 for i in range(len(xs)):
467 xx = [
468 xs[i] + 1,
469 fitted_sorted.shape[1] + fitted_sorted.shape[1] * offset_names,
470 ]
File c:\Users\username\AppData\Local\miniconda3\envs\scFates\lib\site-packages\adjustText\__init__.py:567, in adjust_text(texts, x, y, add_objects, ax, expand_text, expand_points, expand_objects, expand_align, autoalign, va, ha, force_text, force_points, force_objects, lim, precision, only_move, avoid_text, avoid_points, avoid_self, save_steps, save_prefix, save_format, add_step_numbers, *args, **kwargs)
565 ax = ax or plt.gca()
566 r = get_renderer(ax.get_figure())
--> 567 transform = texts[0].get_transform()
568 if (x is not None) & (y is not None):
569 for ix, tupxy in enumerate(zip(x, y)):
IndexError: list index out of rangeFinally, I tried adjusttext==0.7.3 and receive no errors. It would be nice if requirements.txt was updated to pin adjusttext package version accordingly.
LouisFaure
commented
4 months ago This now fixed with scfates v1.0.7, thanks for the notice!
While attempting to plot trends for found clusters:
I encountered the following exception:
In order to workaround the exception, I found the following patch to
adjustText/__init__.pyin theadjustTextpackage allowed me to continue analysis and at least produce plots. The idea is to ensure these four keys are present in theonly_movedict which should prevent the exception, but this is probably a pretty naïve fix.Although I can now produce the plots, they seem to have an issue with the label positioning:
Here is my conda environment: