Merge pull request #20 from genenetwork/feature/biweight-correlation

add biweight r script and tests

3 files changed, 37 insertions, 3 deletions

diff --git a/gn3/computations/biweight.py b/gn3/computations/biweight.py
new file mode 100644
index 0000000..6d031ad
--- /dev/null
+++ b/gn3/computations/biweight.py
@@ -0,0 +1,25 @@
+"""module contains script to call biweight midcorrelation in R"""
+import subprocess
+
+from typing import List
+from typing import Tuple
+
+from gn3.settings import BIWEIGHT_RSCRIPT
+
+
+def calculate_biweight_corr(trait_vals: List,
+                            target_vals: List,
+                            path_to_script: str = BIWEIGHT_RSCRIPT,
+                            command: str = "Rscript"
+                            ) -> Tuple[float, float]:
+    """biweight function"""
+
+    args_1 = ' '.join(str(trait_val) for trait_val in trait_vals)
+    args_2 = ' '.join(str(target_val) for target_val in target_vals)
+    cmd = [command, path_to_script] + [args_1] + [args_2]
+
+    results = subprocess.check_output(cmd, universal_newlines=True)
+
+    (corr_coeff, p_val) = tuple([float(y) for y in results.split()])
+
+    return (corr_coeff, p_val)
diff --git a/gn3/computations/correlations.py b/gn3/computations/correlations.py
index 0fe46ab..bc738a7 100644
--- a/gn3/computations/correlations.py
+++ b/gn3/computations/correlations.py
@@ -7,6 +7,7 @@ from typing import Optional
 from typing import Callable
 
 import scipy.stats
+from gn3.computations.biweight import calculate_biweight_corr
 
 
 def map_shared_keys_to_values(target_sample_keys: List,
@@ -99,8 +100,12 @@ def do_bicor(x_val, y_val) -> Tuple[float, float]:
 package :not packaged in guix
 
     """
-    x_val, y_val = 0, 0
-    return (x_val, y_val)
+
+    try:
+        results = calculate_biweight_corr(x_val, y_val)
+        return results
+    except Exception as error:
+        raise error
 
 
 def filter_shared_sample_keys(this_samplelist,
diff --git a/gn3/settings.py b/gn3/settings.py
index ecfd502..770ba3d 100644
--- a/gn3/settings.py
+++ b/gn3/settings.py
@@ -14,9 +14,13 @@ TMPDIR = os.environ.get("TMPDIR", tempfile.gettempdir())
 RQTL_WRAPPER = "rqtl_wrapper.R"
 
 # SQL confs
-SQL_URI = os.environ.get("SQL_URI", "mysql://webqtlout:webqtlout@localhost/db_webqtl")
+SQL_URI = os.environ.get(
+    "SQL_URI", "mysql://webqtlout:webqtlout@localhost/db_webqtl")
 SECRET_KEY = "password"
 SQLALCHEMY_TRACK_MODIFICATIONS = False
 # gn2 results only used in fetching dataset info
 
 GN2_BASE_URL = "http://www.genenetwork.org/"
+
+# biweight script
+BIWEIGHT_RSCRIPT = "~/genenetwork3/script/calculate_biweight.R"