aboutsummaryrefslogtreecommitdiff
path: root/gn3/computations/correlations.py
diff options
context:
space:
mode:
Diffstat (limited to 'gn3/computations/correlations.py')
-rw-r--r--gn3/computations/correlations.py31
1 files changed, 14 insertions, 17 deletions
diff --git a/gn3/computations/correlations.py b/gn3/computations/correlations.py
index c930df0..8eaa523 100644
--- a/gn3/computations/correlations.py
+++ b/gn3/computations/correlations.py
@@ -49,13 +49,9 @@ def normalize_values(a_values: List,
([2.3, 4.1, 5], [3.4, 6.2, 4.1], 3)
"""
- a_new = []
- b_new = []
for a_val, b_val in zip(a_values, b_values):
if (a_val and b_val is not None):
- a_new.append(a_val)
- b_new.append(b_val)
- return a_new, b_new, len(a_new)
+ yield a_val, b_val
def compute_corr_coeff_p_value(primary_values: List, target_values: List,
@@ -81,8 +77,10 @@ def compute_sample_r_correlation(trait_name, corr_method, trait_vals,
correlation coeff and p value
"""
- (sanitized_traits_vals, sanitized_target_vals,
- num_overlap) = normalize_values(trait_vals, target_samples_vals)
+
+ sanitized_traits_vals, sanitized_target_vals = list(
+ zip(*list(normalize_values(trait_vals, target_samples_vals))))
+ num_overlap = len(sanitized_traits_vals)
if num_overlap > 5:
@@ -114,13 +112,9 @@ def filter_shared_sample_keys(this_samplelist,
filter the values using the shared keys
"""
- this_vals = []
- target_vals = []
for key, value in target_samplelist.items():
if key in this_samplelist:
- target_vals.append(value)
- this_vals.append(this_samplelist[key])
- return (this_vals, target_vals)
+ yield value, this_samplelist[key]
def fast_compute_all_sample_correlation(this_trait,
@@ -139,9 +133,10 @@ def fast_compute_all_sample_correlation(this_trait,
for target_trait in target_dataset:
trait_name = target_trait.get("trait_id")
target_trait_data = target_trait["trait_sample_data"]
- processed_values.append((trait_name, corr_method, *filter_shared_sample_keys(
- this_trait_samples, target_trait_data)))
- with multiprocessing.Pool(4) as pool:
+ processed_values.append((trait_name, corr_method, *list(zip(*list(filter_shared_sample_keys(
+ this_trait_samples, target_trait_data))))
+ ))
+ with multiprocessing.Pool() as pool:
results = pool.starmap(compute_sample_r_correlation, processed_values)
for sample_correlation in results:
@@ -172,8 +167,10 @@ def compute_all_sample_correlation(this_trait,
for target_trait in target_dataset:
trait_name = target_trait.get("trait_id")
target_trait_data = target_trait["trait_sample_data"]
- this_vals, target_vals = filter_shared_sample_keys(
- this_trait_samples, target_trait_data)
+ this_vals, target_vals = list(zip(*list(filter_shared_sample_keys(
+ this_trait_samples, target_trait_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,