diff --git a/WORC/doc/static/user_manual.rst b/WORC/doc/static/user_manual.rst
index 472999b0..32962d75 100644
--- a/WORC/doc/static/user_manual.rst
+++ b/WORC/doc/static/user_manual.rst
@@ -404,7 +404,7 @@ The following outputs and evaluation methods are always generated:
 
 4. The extracted features.
 
-    Stored in the ``Features`` folder, in the files ``features_{featuretoolboxname}_{image_type}_{num}_{sample_id}.hdf5``. Contains a panas series wih the following attributes:
+    Stored in the ``Features`` folder, in the files ``features_{featuretoolboxname}_{image_type}_{num}_{sample_id}.hdf5``. Contains a pandas series with the following attributes:
 
     - feature_labels: the labels or names of the features.
     - feature_values: the value of the features. Each element corresponds with the same element from the feature_labels attribute.
diff --git a/WORC/featureprocessing/StatisticalTestFeatures.py b/WORC/featureprocessing/StatisticalTestFeatures.py
index ab53647c..7e791be8 100644
--- a/WORC/featureprocessing/StatisticalTestFeatures.py
+++ b/WORC/featureprocessing/StatisticalTestFeatures.py
@@ -161,7 +161,10 @@ def StatisticalTestFeatures(features, patientinfo, config, output_csv=None,
             # Optional: perform chi2 test. Only do this when categorical, which we define as less than 20 options.
             unique_values = list(set(fv))
             unique_values.sort()
-            if len(unique_values) == 1:
+            if len(unique_values) == 0: # All NaN
+                print("[WORC Warning] " + fl + " has no value. Replacing chi2 metric value by NaN.")
+                pvalueschi2.append(np.nan)
+            elif len(unique_values) == 1:
                 print("[WORC Warning] " + fl + " has only one value. Replacing chi2 metric value by NaN.")
                 pvalueschi2.append(np.nan)
             elif len(unique_values) <= 20:
diff --git a/WORC/plotting/plot_pvalues_features.py b/WORC/plotting/plot_pvalues_features.py
index a00ef335..20fddf54 100644
--- a/WORC/plotting/plot_pvalues_features.py
+++ b/WORC/plotting/plot_pvalues_features.py
@@ -76,16 +76,21 @@ def manhattan_importance(values, labels, feature_labels,
             ymaxlim = i - 1
             break
 
-    ymin = np.min(values)
+    ymin = np.min(values) # might be zero
+    yposmin = max(np.min(values[values > 0]), np.finfo(values.dtype).eps)
     for i in range(0, 100):
-        if 10**(-i) < ymin:
+        if 10**(-i) < (ymin if ymin > 0 else yposmin):
             yminlim = i
             break
 
     # Set several figure lay-out options
     plt.gca().invert_yaxis()
-    plt.yscale('log')
-    plt.ylim((10**-ymaxlim, 10**-yminlim))
+    if ymin > 0:
+        plt.yscale('log')
+        plt.ylim((10**-ymaxlim, 10**-yminlim))
+    else:
+        plt.yscale('symlog', linthresh=10**-yminlim)
+        plt.ylim((10**-ymaxlim, 0.0))
     plt.xlim((0, max(positions)))
 
     plt.yticks([10**-i for i in range(ymaxlim, yminlim + 1)],