Removed unused gemma code from marker_regression.py and changed the order of a couple functions in collect.py to more accurately reflect when they're called

3 files changed, 15 insertions, 96 deletions

diff --git a/wqflask/wqflask/collect.py b/wqflask/wqflask/collect.py
index 81d03d6c..bf465cdc 100644
--- a/wqflask/wqflask/collect.py
+++ b/wqflask/wqflask/collect.py
@@ -52,7 +52,7 @@ class AnonCollection(object):
         self.id = None
         self.created_timestamp = datetime.datetime.utcnow().strftime('%b %d %Y %I:%M%p')
         self.changed_timestamp = self.created_timestamp #ZS: will be updated when changes are made
-
+        
         #ZS: Find id and set it if the collection doesn't already exist
         if Redis.get(self.key) == "None" or Redis.get(self.key) == None:
             Redis.set(self.key, None) #ZS: For some reason I get the error "Operation against a key holding the wrong kind of value" if I don't do this   
@@ -66,6 +66,7 @@ class AnonCollection(object):
                     collection_exists = True
                     self.id = collection['id']
                     break
+        
         if self.id == None:
             self.id = str(uuid.uuid4())
         
@@ -191,7 +192,19 @@ class UserCollection(object):
         # Probably have to change that
         return redirect(url_for('view_collection', uc_id=uc.id))
 
-
+def process_traits(unprocessed_traits):
+    #print("unprocessed_traits are:", unprocessed_traits)
+    if isinstance(unprocessed_traits, basestring):
+        unprocessed_traits = unprocessed_traits.split(",")
+    traits = set()
+    for trait in unprocessed_traits:
+        #print("trait is:", trait)
+        data, _separator, hmac = trait.rpartition(':')
+        data = data.strip()
+        assert hmac==user_manager.actual_hmac_creation(data), "Data tampering?"
+        traits.add                                                                                               (str(data))
+    return traits
+        
 def report_change(len_before, len_now):
     new_length = len_now - len_before
     if new_length:
@@ -224,7 +237,6 @@ def collections_add():
                                 collections = collection_names,
                               )
 
-
 @app.route("/collections/new")
 def collections_new():
     params = request.args
@@ -248,20 +260,6 @@ def collections_new():
     else:
         CauseAnError
 
-
-def process_traits(unprocessed_traits):
-    #print("unprocessed_traits are:", unprocessed_traits)
-    if isinstance(unprocessed_traits, basestring):
-        unprocessed_traits = unprocessed_traits.split(",")
-    traits = set()
-    for trait in unprocessed_traits:
-        #print("trait is:", trait)
-        data, _separator, hmac = trait.rpartition(':')
-        data = data.strip()
-        assert hmac==user_manager.actual_hmac_creation(data), "Data tampering?"
-        traits.add                                                                                               (str(data))
-    return traits
-
 def create_new(collection_name):
     params = request.args
     
diff --git a/wqflask/wqflask/marker_regression/marker_regression.py b/wqflask/wqflask/marker_regression/marker_regression.py
index f340e309..7ad5d25a 100644
--- a/wqflask/wqflask/marker_regression/marker_regression.py
+++ b/wqflask/wqflask/marker_regression/marker_regression.py
@@ -309,49 +309,6 @@ class MarkerRegression(object):
                 perm_results = self.perm_output,
             )
 
-
-    def run_gemma(self):
-        """Generates p-values for each marker using GEMMA"""
-
-        self.gen_pheno_txt_file()
-
-        #os.chdir("/home/zas1024/gene/web/gemma")
-
-        gemma_command = GEMMA_COMMAND + ' -bfile %s -k output_%s.cXX.txt -lmm 1 -o %s_output' % (
-                                                                                                 self.dataset.group.name,
-                                                                                                 self.dataset.group.name,
-                                                                                                 self.dataset.group.name)
-        #logger.debug("gemma_command:" + gemma_command)
-
-        os.system(gemma_command)
-
-        included_markers, p_values = self.parse_gemma_output()
-
-        self.dataset.group.get_specified_markers(markers = included_markers)
-        self.dataset.group.markers.add_pvalues(p_values)
-        return self.dataset.group.markers.markers
-
-    def parse_gemma_output(self):
-        included_markers = []
-        p_values = []
-        # Use a temporary file name here!
-        with open(webqtlConfig.GENERATED_TEXT_DIR+"/{}_output.assoc.txt".format(self.dataset.group.name)) as output_file:
-            for line in output_file:
-                if line.startswith("chr"):
-                    continue
-                else:
-                    included_markers.append(line.split("\t")[1])
-                    p_values.append(float(line.split("\t")[10]))
-                    #p_values[line.split("\t")[1]] = float(line.split("\t")[10])
-        #logger.debug("p_values: ", p_values)
-        return included_markers, p_values
-
-    def gen_pheno_txt_file(self):
-        """Generates phenotype file for GEMMA"""
-        with open(webqtlConfig.GENERATED_TEXT_DIR+"{}.fam".format(self.dataset.group.name), "w") as outfile:
-            for i, sample in enumerate(self.samples):
-                outfile.write(str(sample) + " " + str(sample) + " 0 0 0 " + str(self.vals[i]) + "\n")
-
     def run_rqtl_plink(self):
         # os.chdir("") never do this inside a webserver!!
 
@@ -425,41 +382,6 @@ class MarkerRegression(object):
 
         output_file.close()
 
-    def gen_pheno_txt_file_rqtl(self, pheno_filename = ''):
-        ped_sample_list = self.get_samples_from_ped_file()
-        output_file = open("%s%s.txt" % (TMPDIR, pheno_filename), "wb")
-        header = 'FID\tIID\t%s\n' % self.this_trait.name
-        output_file.write(header)
-
-        new_value_list = []
-
-        #if valueDict does not include some strain, value will be set to -9999 as missing value
-        for i, sample in enumerate(ped_sample_list):
-            try:
-                value = self.vals[i]
-                value = str(value).replace('value=','')
-                value = value.strip()
-            except:
-                value = -9999
-
-            new_value_list.append(value)
-
-
-        new_line = ''
-        for i, sample in enumerate(ped_sample_list):
-            j = i+1
-            value = new_value_list[i]
-            new_line += '%s\t%s\t%s\n'%(sample, sample, value)
-
-            if j%1000 == 0:
-                output_file.write(newLine)
-                new_line = ''
-
-        if new_line:
-            output_file.write(new_line)
-
-        output_file.close()
-
     # get strain name from ped file in order
     def get_samples_from_ped_file(self):
         ped_file= open("{}/{}.ped".format(PLINK_PATH, self.dataset.group.name),"r")
diff --git a/wqflask/wqflask/user_manager.py b/wqflask/wqflask/user_manager.py
index 598af0a6..1e831896 100644
--- a/wqflask/wqflask/user_manager.py
+++ b/wqflask/wqflask/user_manager.py
@@ -158,7 +158,6 @@ def verify_cookie(cookie):
     assert the_signature == actual_hmac_creation(the_uuid), "Uh-oh, someone tampering with the cookie?"
     return the_uuid
 
-
 def create_signed_cookie():
     the_uuid = str(uuid.uuid4())
     signature = actual_hmac_creation(the_uuid)