fix nan handling

dwnorton · Jul 29, 2020 · c8be573 · c8be573
1 parent 3b26ccf
commit c8be573
Showing 1 changed file with 3 additions and 4 deletions.
diff --git a/scanpy/preprocessing/_highly_variable_genes.py b/scanpy/preprocessing/_highly_variable_genes.py
@@ -120,13 +120,12 @@ def _highly_variable_genes_seurat_v3(
 
     # this is done in SelectIntegrationFeatures() in Seurat v3
     ranked_norm_gene_vars = ranked_norm_gene_vars.astype(np.float32)
-    ranked_norm_gene_vars[ranked_norm_gene_vars >= n_top_genes] = np.nan
-    ranked_norm_gene_vars = np.ma.masked_invalid(ranked_norm_gene_vars)
-    median_ranked = np.median(ranked_norm_gene_vars, axis=0)
     num_batches_high_var = np.sum(
         (ranked_norm_gene_vars < n_top_genes).astype(int), axis=0
     )
-    ranked_norm_gene_vars = np.ma.filled(ranked_norm_gene_vars, fill_value=np.nan)
+    ranked_norm_gene_vars[ranked_norm_gene_vars >= n_top_genes] = np.nan
+    ma_ranked = np.ma.masked_invalid(ranked_norm_gene_vars)
+    median_ranked = np.ma.median(ma_ranked, axis=0).filled(np.nan)
 
     df = pd.DataFrame(index=np.array(adata.var_names))
     df['highly_variable_nbatches'] = num_batches_high_var