Does stlearn support customly defined celltype in trajectory analysis?

[QuinceyLv]

Hello, I'm trying to do trajectory analysis with stlearn. When I looked into stlearn/plotting/trajectory/pseudotime_plot.py, line 130: https://github.com/BiomedicalMachineLearning/stLearn/blob/e22bc1fa50c91667a7f364f2091155f5e4633610/stlearn/plotting/trajectory/pseudotime_plot.py#L130

It seems that I can't use customly defined celltype (e.g. Celltype_A, Celltype_B).

So what should I do if I hope to use celltypes?

Thanks in advance.

[duypham2108]

I will try to fix this problem (poorly designed function before). For now, I think you can create another column and change your custom cell type to this type:

adata.obs["celltype_num"] = adata.obs["celltype"].copy()
num_cat = list(range(0,len(adata.obs["celltype_num"].cat.categories)))
adata.obs["celltype_num"].cat.categories = list(map(lambda x: str(x),num_cat))

Then you can run with the newly created label

[QuinceyLv]

I will try to fix this problem (poorly designed function before). For now, I think you can create another column and change your custom cell type to this type:

adata.obs["celltype_num"] = adata.obs["celltype"].copy()
num_cat = list(range(0,len(adata.obs["celltype_num"].cat.categories)))
adata.obs["celltype_num"].cat.categories = list(map(lambda x: str(x),num_cat))

Then you can run with the newly created label

Thanks for reply. I tried the code and got an error:

adata_SME.obs["celltype_num"] = adata_SME.obs["celltype"].copy() num_cat = list(range(0,len(adata_SME.obs["celltype_num"].cat.categories))) num_cat [0, 1, 2, 3, 4, 5] adata_SME.obs["celltype_num"].cat.categories = list(map(lambda x: str(x),num_cat)) Traceback (most recent call last): File "", line 1, in File "foo/bar/envs/stlearn/lib/python3.8/site-packages/pandas/core/base.py", line 178, in setattr object.setattr(self, key, value) File "foo/bar/envs/stlearn/lib/python3.8/site-packages/pandas/core/accessor.py", line 99, in _setter return self._delegate_property_set(name, new_values) File "foo/bar/envs/stlearn/lib/python3.8/site-packages/pandas/core/arrays/categorical.py", line 2460, in _delegate_property_set return setattr(self._parent, name, new_values) AttributeError: can't set attribute

I'm using stlearn == 0.4.12 and pandas == 2.0.1.

Besides, if I run with label "celltype_num", how could I put corresponding celltype label on plots instead of the numbers?

[duypham2108]

Can you set adata_SME.obs["celltype_num"].cat.categories to the list of numbers (string type) that corresponds to the number of your cluster?

It will be a problem if you want to plot directly the label. I will try to fix it soon, hopefully within this week. A quick solution is that you export to pdf and replace the number with your label by illustrator.

[QuinceyLv]

Can you set adata_SME.obs["celltype_num"].cat.categories to the list of numbers (string type) that corresponds to the number of your cluster?

It will be a problem if you want to plot directly the label. I will try to fix it soon, hopefully within this week. A quick solution is that you export to pdf and replace the number with your label by illustrator.

Great help. Really appreciated.

[QuinceyLv]

@duypham2108 Hello, I tried to plot trajectory tree using st.pl.trajectory.tree_plot():

st.pl.trajectory.tree_plot(adata_SME, use_label="leiden")

I expected to get plots like the tutorial:

While I got:

[duypham2108]

We removed that option. You can downgrade to the version 0.4.8 to plot it. We may bring it back in the next version.