"InvalidIndexError: Reindexing only valid with uniquely valued Index objects" when running st.tl.cci.run with custom LR set

[je71xusa]

liana_consensus_db.csv
Hi,

I am trying to run CCI with a custom LR database from liana (see above).

I load it like this:

# CCI
df = pd.read_csv('liana_consensus_db.csv')
lrs = np.array(df['x'], dtype='<U18')
print(len(lrs))
lrs

and that gives me this output:

3998
array(['Dll1_Notch1', 'Dll1_Notch2', 'Dll1_Notch4', ..., 'Serpina1c_Lrp1',
 'Serpina1d_Lrp1', 'Serpina1e_Lrp1'], dtype='<U18')

But when I run this:

st.tl.cci.run(adata, lrs,
 min_spots = 20, #Filter out any LR pairs with no scores for less than min_spots
 distance=None, # None defaults to spot+immediate neighbours; distance=0 for within-spot mode
 n_pairs=10000 # Number of random pairs to generate, recommend ~10,000
 #n_cpus=4, # Number of CPUs for parallel. If None, detects & use all available.
 )

I get this huge error indicating that pd.concat did not work for lr_features and quant_df in perm_utils.py

---------------------------------------------------------------------------
InvalidIndexError Traceback (most recent call last)
/tmp/ipykernel_4189510/1130224459.py in <module>
 2 min_spots = 20, #Filter out any LR pairs with no scores for less than min_spots
 3 distance=None, # None defaults to spot+immediate neighbours; distance=0 for within-spot mode
----> 4 n_pairs=10000 # Number of random pairs to generate, recommend ~10,000
 5 #n_cpus=4, # Number of CPUs for parallel. If None, detects & use all available.
 6 )

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/stlearn/tools/microenv/cci/analysis.py in run(adata, lrs, min_spots, distance, n_pairs, n_cpus, use_label, adj_method, pval_adj_cutoff, min_expr, save_bg, neg_binom, verbose)
 347 verbose,
 348 save_bg=save_bg,
--> 349 neg_binom=neg_binom,
 350 )
 351 

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/stlearn/tools/microenv/cci/permutation.py in perform_spot_testing(adata, lr_scores, lrs, n_pairs, neighbours, het_vals, min_expr, adj_method, pval_adj_cutoff, verbose, save_bg, neg_binom, quantiles)
 57 ####### Quantiles to select similar gene to LRs to gen. rand-pairs #######
 58 lr_expr = adata[:, lr_genes].to_df()
---> 59 lr_feats = get_lr_features(adata, lr_expr, lrs, quantiles)
 60 l_quants = lr_feats.loc[
 61 lrs, [col for col in lr_feats.columns if "L_" in col]

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/stlearn/tools/microenv/cci/perm_utils.py in get_lr_features(adata, lr_expr, lrs, quantiles)
 323 ]
 324 quant_df = pd.DataFrame(lr_quants, columns=lr_cols, index=lrs)
--> 325 lr_features = pd.concat((lr_features, quant_df), axis=1)
 326 adata.uns["lrfeatures"] = lr_features
 327 

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/pandas/util/_decorators.py in wrapper(*args, **kwargs)
 309 stacklevel=stacklevel,
 310 )
--> 311 return func(*args, **kwargs)
 312 
 313 return wrapper

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/pandas/core/reshape/concat.py in concat(objs, axis, join, ignore_index, keys, levels, names, verify_integrity, sort, copy)
 305 )
 306 
--> 307 return op.get_result()
 308 
 309 

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/pandas/core/reshape/concat.py in get_result(self)
 526 obj_labels = obj.axes[1 - ax]
 527 if not new_labels.equals(obj_labels):
--> 528 indexers[ax] = obj_labels.get_indexer(new_labels)
 529 
 530 mgrs_indexers.append((obj._mgr, indexers))

~/anaconda3/envs/stlearn_Test/lib/python3.7/site-packages/pandas/core/indexes/base.py in get_indexer(self, target, method, limit, tolerance)
 3440 
 3441 if not self._index_as_unique:
-> 3442 raise InvalidIndexError(self._requires_unique_msg)
 3443 
 3444 if not self._should_compare(target) and not is_interval_dtype(self.dtype):

InvalidIndexError: Reindexing only valid with uniquely valued Index objects

Using lrs = st.tl.cci.load_lrs(['connectomeDB2020_lit'], species='mouse') works but I do not see the difference in formatting between that and the liana dataset. Can you help me?

Thank you!

[paularstrpo]

I run into the same issue when specifying a custom LR list based on CellChatDB. As a related question, does stlearn support custom LR inputs?