diff --git a/spateo/tools/CCI_effects_modeling/MuSIC_downstream.py b/spateo/tools/CCI_effects_modeling/MuSIC_downstream.py
index 5fff7c3b..03d16957 100644
--- a/spateo/tools/CCI_effects_modeling/MuSIC_downstream.py
+++ b/spateo/tools/CCI_effects_modeling/MuSIC_downstream.py
@@ -539,7 +539,7 @@ def compute_and_visualize_diagnostics(
                 data=df,
                 x="Gene",
                 y="Pearson coefficient",
-                palette=colors["Pearson coefficient"],
+                # palette=colors["Pearson coefficient"],
                 edgecolor="black",
                 dodge=True,
             )
@@ -567,7 +567,7 @@ def compute_and_visualize_diagnostics(
                 data=df,
                 x="Gene",
                 y="Spearman coefficient",
-                palette=colors["Spearman coefficient"],
+                # palette=colors["Spearman coefficient"],
                 edgecolor="black",
                 dodge=True,
             )
@@ -593,7 +593,7 @@ def compute_and_visualize_diagnostics(
                 data=df,
                 x="Gene",
                 y="Pearson coefficient (expressing cells)",
-                palette=colors["Pearson coefficient (expressing cells)"],
+                # palette=colors["Pearson coefficient (expressing cells)"],
                 edgecolor="black",
                 dodge=True,
             )
@@ -619,7 +619,7 @@ def compute_and_visualize_diagnostics(
                 data=df,
                 x="Gene",
                 y="Spearman coefficient (expressing cells)",
-                palette=colors["Spearman coefficient (expressing cells)"],
+                # palette=colors["Spearman coefficient (expressing cells)"],
                 edgecolor="black",
                 dodge=True,
             )
@@ -4173,8 +4173,8 @@ def cell_type_specific_interactions(
             vmin = 0
             vmax = 1 if normalize else df.max().max()
 
-            fig, ax = plt.subplots(nrows=1, ncols=1, figsize=figsize)
-            divider = make_axes_locatable(ax)
+            fig, axes = plt.subplots(nrows=1, ncols=1, figsize=figsize)
+            divider = make_axes_locatable(axes)
             ax2 = divider.append_axes("right", size=ax2_size, pad=0)
 
             # Keep track of groups:
@@ -4212,7 +4212,7 @@ def cell_type_specific_interactions(
                 vmin=vmin,
                 vmax=vmax,
                 mask=mask,
-                ax=ax,
+                ax=axes,
             )
 
             # Outer frame:
@@ -4221,7 +4221,7 @@ def cell_type_specific_interactions(
                 spine.set_linewidth(thickness * 2.5)
 
             # Adjust colorbar settings:
-            divider = make_axes_locatable(ax)
+            divider = make_axes_locatable(axes)
             # Append axes to the top of the plot, where the colorbar will be placed
             if df.shape[0] > df.shape[1]:
                 cax = divider.append_axes("top", size="30%", pad=0)
@@ -4236,11 +4236,11 @@ def cell_type_specific_interactions(
             cbar.ax.tick_params(labelsize=fontsize * 1.5)
             cbar.ax.set_aspect(0.02)
 
-            ax.set_xlabel(x_label, fontsize=fontsize * 1.25)
-            ax.set_ylabel("Cell Type-Specific Target", fontsize=fontsize * 1.25)
-            ax.tick_params(axis="x", labelsize=fontsize, rotation=90)
-            ax.tick_params(axis="y", labelsize=fontsize)
-            ax.set_title(title, fontsize=fontsize * 1.5, pad=20)
+            axes.set_xlabel(x_label, fontsize=fontsize * 1.25)
+            axes.set_ylabel("Cell Type-Specific Target", fontsize=fontsize * 1.25)
+            axes.tick_params(axis="x", labelsize=fontsize, rotation=90)
+            axes.tick_params(axis="y", labelsize=fontsize)
+            axes.set_title(title, fontsize=fontsize * 1.5, pad=20)
 
             # Use the saved name for the AnnData object to define part of the name of the saved file:
             base_name = os.path.basename(self.adata_path)
@@ -6908,7 +6908,6 @@ def CCI_deg_detection_setup(
                 if scipy.sparse.issparse(adata.X):
                     nnz_counts = np.array(adata[:, all_TFs].X.getnnz(axis=0)).flatten()
                 else:
-                    # nnz_counts = np.array(adata[:, all_TFs].X.getnnz(axis=0)).flatten()
                     nnz_counts = np.count_nonzero(
                         adata[:, all_TFs].X, axis=0
                     ).flatten()  # np.ndarray have no getnnz attributes
@@ -7637,7 +7636,6 @@ def deg_effect_heatmap(
                         all_cells_affected = dm[dm[f"regulator_{interaction}"] > 0]
                     else:
                         all_cells_affected = dm[dm[interaction] > 0]
-
                     specificity = (all_coeffs_target.loc[all_cells_affected.index, interaction] != 0).mean()
                     all_plot_values.loc[interaction, target] = specificity
         all_plot_values.index = [replace_col_with_collagens(f) for f in all_plot_values.index]
diff --git a/spateo/tools/cluster/leiden.py b/spateo/tools/cluster/leiden.py
index 8846b8ff..5b1e7fcd 100644
--- a/spateo/tools/cluster/leiden.py
+++ b/spateo/tools/cluster/leiden.py
@@ -5,6 +5,7 @@
 import leidenalg
 import numpy as np
 import scipy
+import scipy.sparse
 from sklearn.neighbors import kneighbors_graph
 
 from ...configuration import SKM
@@ -87,6 +88,8 @@ def calculate_leiden_partition(
     if adj is not None:
         if isinstance(adj, np.ndarray):
             G = adj_to_igraph(adj.tolist())
+        elif scipy.sparse.issparse(adj):
+            G = adj_to_igraph(adj.A.tolist())
         else:
             G = igraph.Graph(n=adj.shape[0])
             for i, j in zip(*adj.nonzero()):
diff --git a/spateo/tools/cluster/utils.py b/spateo/tools/cluster/utils.py
index 57c5e976..25e6bf8b 100644
--- a/spateo/tools/cluster/utils.py
+++ b/spateo/tools/cluster/utils.py
@@ -295,7 +295,8 @@ def spatial_adj(
     _, adata = neighbors(
         adata,
         n_neighbors=s_neigh,
-        basis=spatial_key,
+        basis="spatial",
+        spatial_key=spatial_key,
         n_pca_components=n_pca_components,
     )