Update feature description

spatialprofilingtoolbox/db/data_model/feature_descriptions.tsv

@@ -1,7 +1,7 @@
 spatial autocorrelation	For a given cell phenotype (first specifier), i.e. cell set, the p-value associated with the Moran I statistic value for this cell set, with respect to the standard null distribution for this statistic.
 neighborhood enrichment	For two given cell phenotypes (first and second specifiers), the estimated p-value (CDF of z-score) of the number of occurrences of neighbor relations between two cells of the given respective phenotypes, the z-score computed in reference to a bootstrapped empirical distribution constructed using 1000 random permutations of the cell set with labels fixed.
 co-occurrence	For two given cell phenotypes (first and second specifiers), and a given radius (third specifier), the ratio of the two probabilities of occurrence of the first phenotype within the given radius around some cell, calculated with and without conditioning on at least one occurrence of the second phenotype in the same radius.
-ripley	For a given cell phenotype (first specifier), the p-value associated with the Ripley F-statistic for the point set of cell locations, at the radius scale for which this p-value is smallest.
+ripley	For a given cell phenotype (first specifier), the radius scale which is the expected value with respect to the p-values associated with elevated Ripley F-statistic for the point set of cell locations, that is, a radius scale at which non-trivial clustering is observed in the Ripley F sense.
 gnn importance score	For a given cohort stratification variable (the specifier), the integer rank of each cell (the subjects of the feature) with respect to the importance scores derived from a GNN trained on this variable.
 population fractions	For a given cell phenotype, the average number of cells of that phenotype in the given sample relative to the number of cells in the sample.
 proximity	For a given cell phenotype (first specifier), the average number of cells of a second phenotype (second specifier) within a specified radius (third specifier).

spatialprofilingtoolbox/ondemand/worker.py

@@ -86,6 +86,7 @@ def _compute(self, job: Job) -> None:
         except Exception as error:
             logger.error(error)
             print_exception(type(error), error, error.__traceback__)
+            self._no_value_wrapup(job)
         finally:
             generic_handler.disalarm()
 

spatialprofilingtoolbox/workflow/common/squidpy.py

@@ -143,6 +143,7 @@ def ripley_custom(
 
     pvalues /= n_simulations + 1
     # pvalues = np.minimum(pvalues, 1 - pvalues)
+    pvalues = 1 - pvalues
 
     obs_df = _reshape_res(obs_arr.T, columns=le.classes_, index=bins, var_name=cluster_key)
     sims_df = _reshape_res(sims.T, columns=np.arange(n_simulations), index=bins, var_name="simulations")
@@ -230,7 +231,7 @@ def _summarize_ripley(unstructured_metrics: dict[str, NDArray[Any]]) -> float |
     total_p = np_sum(pvalues)
     if len(bins) != len(pvalues) or total_p == 0:
         return None
-    return inner(pvalues, bins) / np_sum(pvalues)
+    return round(inner(pvalues, bins) / total_p, 1)
 
 
 def _summarize_spatial_autocorrelation(unstructured_metrics: DataFrame) -> float | None: