spc.est_spectra returns error "AttributeError: 'Series' object has no attribute 'nonzero'"

[halterc]

Hi

when I try to run the example from the tutorial locally (the online colab crashes due to running out of memory), I get the following error when I try to run:

model = spc.est_spectra(adata=adata, gene_set_dictionary=annotations, use_highly_variable=True, cell_type_key="cell_type_annotations", use_weights=True, lam=0.1, # varies depending on data and gene sets, try between 0.5 and 0.001 delta=0.001, kappa=None, rho=0.001, use_cell_types=True, n_top_vals=50, label_factors=True, overlap_threshold=0.2, clean_gs = True, min_gs_num = 3, num_epochs=2 #here running only 2 epochs for time reasons, we recommend 10,000 epochs for most datasets )

---

AttributeError Traceback (most recent call last)
/var/folders/tw/3_vqzv_517515r5vv8pnrljc0000gn/T/ipykernel_10048/2900975804.py in ?()
----> 1 model = spc.est_spectra(adata=bdata,
2 gene_set_dictionary=bnnotations,
3 use_highly_variable=True,
4 cell_type_key="cell_type_annotations",

/opt/homebrew/anaconda3/lib/python3.11/site-packages/Spectra/Spectra.py in ?(adata, gene_set_dictionary, L, use_highly_variable, cell_type_key, use_weights, lam, delta, kappa, rho, use_cell_types, n_top_vals, filter_sets, label_factors, clean_gs, min_gs_num, overlap_threshold, **kwargs)
1534 if use_highly_variable:
1535 idx_to_use = (
1536 bools | adata.var.highly_variable
1537 ) # take union of highly variable and gene set genes (todo: add option to change this at some point)
-> 1538 X = adata.X[:, idx_to_use]
1539 vocab = adata.var_names[idx_to_use]
1540 adata.var["spectra_vocab"] = idx_to_use
1541 else:

/opt/homebrew/anaconda3/lib/python3.11/site-packages/scipy/sparse/_index.py in ?(self, key)
29 def getitem(self, key):
---> 30 index, new_shape = self._validate_indices(key)
31
32 # 1D array
33 if len(index) == 1:

/opt/homebrew/anaconda3/lib/python3.11/site-packages/scipy/sparse/_index.py in ?(self, key)
265 if ix.shape != mid_shape:
266 raise IndexError(
267 f"bool index {i} has shape {mid_shape} instead of {ix.shape}"
268 )
--> 269 index.extend(ix.nonzero())
270 array_indices.extend(range(index_ndim, tmp_ndim))
271 index_ndim = tmp_ndim
272 else: # dense array

/opt/homebrew/anaconda3/lib/python3.11/site-packages/pandas/core/generic.py in ?(self, name)
6295 and name not in self._accessors
6296 and self._info_axis._can_hold_identifiers_and_holds_name(name)
6297 ):
6298 return self[name]
-> 6299 return object.getattribute(self, name)

AttributeError: 'Series' object has no attribute 'nonzero'

Seems adata.var is always a pandas dataframe, as this returns an error
adata.var = adata.var.to_numpy()

Am I missing something?

Thanks for your help.

[TBar123]

Hey- I also had this problem! Fixed it by downgrading my scipy to version 1.14.1.

[Almalvas]

Try adata.X = adata.X.toarray() if adata.X is a sparse matrix. This helped me.