bivariate: clean code; yaml: change defaults: spearman

src/dimet/config/analysis/method/bivariate_analysis.yaml

@@ -6,16 +6,16 @@ name: Computation of the correlation of MDV profiles, or the metabolite time cou
 # (**) : automatically will run
 
 conditions_MDV_comparison: # (**) if >= 2 conditions and >=1 timepoint (timepoints run separately)
-  isotopologue_proportions: pearson
+  isotopologue_proportions: spearman
 
 timepoints_MDV_comparison:  # (**) if >= 1 condition and >=2 timepoints
-  isotopologue_proportions: pearson
+  isotopologue_proportions: spearman
 
 conditions_metabolite_time_profiles:  # (**) if >= 2 conditions  AND >=2 time points in data
-  abundances: pearson
-  mean_enrichment: pearson
+  abundances: spearman
+  mean_enrichment: spearman
 
-# test modification doable with external config 'statistical_test'
+# test modification accessible through external config 'statistical_test'
 
 correction_method: fdr_bh
 

src/dimet/method/__init__.py

@@ -1,7 +1,7 @@
 import logging
 import os
 import sys
-from typing import Union
+from typing import Dict, Union
 
 import hydra
 from omegaconf import DictConfig, ListConfig, OmegaConf, open_dict
@@ -191,6 +191,14 @@ class BivariateAnalysisConfig(MethodConfig):
     """
     correction_method: str = "fdr_bh"
     output_include_gmean_arr_columns: bool = True
+    conditions_MDV_comparison: Dict[str, str] = {
+        'isotopologue_proportions': 'spearman'}
+    timepoints_MDV_comparison: Dict[str, str] = {
+        'isotopologue_proportions': 'spearman'}
+    conditions_metabolite_time_profiles: Dict[str, str] = {
+        'abundances': 'spearman',
+        'mean_enrichment': 'spearman'
+    }
 
     def build(self) -> "BivariateAnalysis":
         return BivariateAnalysis(config=self)

src/dimet/processing/bivariate_analysis.py

@@ -32,7 +32,7 @@ def compute_statistical_correlation(df: pd.DataFrame,
     stat_list = []
     pvalue_list = []
     for i, metabolite in enumerate(list(df['metabolite'])):
-        # array of n-(timepoints or m+x) geometrical means values
+        # array of n (timepoints or m+x) geometrical means values
         array_1 = df.loc[metabolite, "gmean_arr_1"]
         array_2 = df.loc[metabolite, "gmean_arr_2"]
         if test == "pearson":
@@ -79,7 +79,7 @@ def compute_bivariate_by_behavior(
     """
     performs two steps:
     1. calls functions to compute geometric means, obtaining df's inside dict
-    2. computes the bivariate statistical test (pearson by default)
+    2. computes the bivariate statistical test
     """
     if behavior == "conditions_MDV_comparison":
         df_dict = conditions_MDV_gmean_df_dict(df, metadata_df, comparison)