From 95eb105421d7fbe0b0ebf1de2f89a558eecbf0da Mon Sep 17 00:00:00 2001
From: Alexander Kabui
Date: Wed, 11 Aug 2021 09:18:19 +0300
Subject: use normal function for correlation  (#34)

* use normal function for correlation + rename functions

* update test for sample correlation

* use normal function for tissue correlation + rename functions---
 gn3/computations/correlations.py | 26 +++++++++++++++-----------
 1 file changed, 15 insertions(+), 11 deletions(-)

(limited to 'gn3/computations')

diff --git a/gn3/computations/correlations.py b/gn3/computations/correlations.py
index 56f483c..1fd3213 100644
--- a/gn3/computations/correlations.py
+++ b/gn3/computations/correlations.py
@@ -124,11 +124,12 @@ def filter_shared_sample_keys(this_samplelist,
     return (this_vals, target_vals)
 
 
-def compute_all_sample_correlation(this_trait,
-                                   target_dataset,
-                                   corr_method="pearson") -> List:
+def speed_compute_all_sample_correlation(this_trait,
+                                         target_dataset,
+                                         corr_method="pearson") -> List:
     """Given a trait data sample-list and target__datasets compute all sample
     correlation
+    this functions uses multiprocessing if not use the normal fun
 
     """
     # xtodo fix trait_name currently returning single one
@@ -160,9 +161,9 @@ def compute_all_sample_correlation(this_trait,
         key=lambda trait_name: -abs(list(trait_name.values())[0]["corr_coefficient"]))
 
 
-def benchmark_compute_all_sample(this_trait,
-                                 target_dataset,
-                                 corr_method="pearson") -> List:
+def compute_all_sample_correlation(this_trait,
+                                   target_dataset,
+                                   corr_method="pearson") -> List:
     """Temp function to benchmark with compute_all_sample_r alternative to
     compute_all_sample_r where we use multiprocessing
 
@@ -174,6 +175,7 @@ def benchmark_compute_all_sample(this_trait,
         target_trait_data = target_trait["trait_sample_data"]
         this_vals, target_vals = filter_shared_sample_keys(
             this_trait_samples, target_trait_data)
+
         sample_correlation = compute_sample_r_correlation(
             trait_name=trait_name,
             corr_method=corr_method,
@@ -190,7 +192,9 @@ def benchmark_compute_all_sample(this_trait,
             "num_overlap": num_overlap
         }
         corr_results.append({trait_name: corr_result})
-    return corr_results
+    return sorted(
+        corr_results,
+        key=lambda trait_name: -abs(list(trait_name.values())[0]["corr_coefficient"]))
 
 
 def tissue_correlation_for_trait(
@@ -336,7 +340,7 @@ def compute_all_lit_correlation(conn, trait_lists: List,
     return sorted_lit_results
 
 
-def compute_all_tissue_correlation(primary_tissue_dict: dict,
+def compute_tissue_correlation(primary_tissue_dict: dict,
                                    target_tissues_data: dict,
                                    corr_method: str):
     """Function acts as an abstraction for tissue_correlation_for_trait\
@@ -382,9 +386,9 @@ def process_trait_symbol_dict(trait_symbol_dict, symbol_tissue_vals_dict) -> Lis
     return traits_tissue_vals
 
 
-def compute_tissue_correlation(primary_tissue_dict: dict,
-                               target_tissues_data: dict,
-                               corr_method: str):
+def speed_compute_tissue_correlation(primary_tissue_dict: dict,
+                                     target_tissues_data: dict,
+                                     corr_method: str):
     """Experimental function that uses multiprocessing for computing tissue
     correlation
 
-- 
cgit v1.2.3