Merge branch 'testing' of https://github.com/genenetwork/genenetwork2 into production

1 files changed, 18 insertions, 214 deletions

diff --git a/wqflask/base/trait.py b/wqflask/base/trait.py
index acc055d8..3daf9ea9 100644
--- a/wqflask/base/trait.py
+++ b/wqflask/base/trait.py
@@ -18,17 +18,13 @@ import simplejson as json
 from MySQLdb import escape_string as escape
 from pprint import pformat as pf
 
-from flask import Flask, g, request
+from flask import Flask, g, request, url_for
 
 from utility.logger import getLogger
 logger = getLogger(__name__ )
 
 from wqflask import user_manager
 
-def print_mem(stage=""):
-    mem = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
-    print("{}: {}".format(stage, mem/1024))
-
 class GeneralTrait(object):
     """
     Trait class defines a trait in webqtl, can be either Microarray,
@@ -63,9 +59,7 @@ class GeneralTrait(object):
         self.symbol = None
 
         self.LRS_score_repr = "N/A"
-        self.LRS_score_value = 0
         self.LRS_location_repr = "N/A"
-        self.LRS_location_value = 1000000
 
         if kw.get('fullname'):
             name2 = value.split("::")
@@ -82,90 +76,6 @@ class GeneralTrait(object):
             if get_sample_info != False:
                 self = retrieve_sample_data(self, self.dataset)
 
-
-    def get_name(self):
-        stringy = ""
-        if self.dataset and self.name:
-            stringy = "%s::%s" % (self.dataset, self.name)
-            if self.cellid:
-                stringy += "::" + self.cellid
-        else:
-            stringy = self.description
-        return stringy
-
-
-    def get_given_name(self):
-        """
-         when user enter a trait or GN generate a trait, user want show the name
-         not the name that generated by GN randomly, the two follow function are
-         used to give the real name and the database. displayName() will show the
-         database also, getGivenName() just show the name.
-         For other trait, displayName() as same as getName(), getGivenName() as
-         same as self.name
-
-         Hongqiang 11/29/07
-
-        """
-        stringy = self.name
-        if self.dataset and self.name:
-            desc = self.dataset.get_desc()
-            if desc:
-                #desc = self.handle_pca(desc)
-                stringy = desc
-        return stringy
-
-
-    def display_name(self):
-        stringy = ""
-        if self.dataset and self.name:
-            desc = self.dataset.get_desc()
-            #desc = self.handle_pca(desc)
-            if desc:
-                #desc = self.handle_pca(desc)
-                #stringy = desc
-                #if desc.__contains__('PCA'):
-                #    desc = desc[desc.rindex(':')+1:].strip()
-                #else:
-                #    desc = desc[:desc.index('entered')].strip()
-                #desc = self.handle_pca(desc)
-                stringy = "%s::%s" % (self.dataset, desc)
-            else:
-                stringy = "%s::%s" % (self.dataset, self.name)
-                if self.cellid:
-                    stringy += "::" + self.cellid
-        else:
-            stringy = self.description
-
-        return stringy
-
-
-    #def __str__(self):
-    #       #return "%s %s" % (self.getName(), self.group)
-    #       return self.getName()
-    #__str__ = getName
-    #__repr__ = __str__
-
-    def export_data(self, samplelist, the_type="val"):
-        """
-        export data according to samplelist
-        mostly used in calculating correlation
-
-        """
-        result = []
-        for sample in samplelist:
-            if self.data.has_key(sample):
-                if the_type=='val':
-                    result.append(self.data[sample].val)
-                elif the_type=='var':
-                    result.append(self.data[sample].var)
-                elif the_type=='N':
-                    result.append(self.data[sample].N)
-                else:
-                    raise KeyError, `the_type`+' the_type is incorrect.'
-            else:
-                result.append(None)
-        return result
-
     def export_informative(self, include_variance=0):
         """
         export informative sample
@@ -185,19 +95,6 @@ class GeneralTrait(object):
                     sample_aliases.append(sample_data.name2)
         return  samples, vals, the_vars, sample_aliases
 
-
-    @property
-    def name_header_fmt(self):
-        '''Return a human-readable name for use in page header'''
-        if self.dataset.type == 'ProbeSet':
-            return self.symbol
-        elif self.dataset.type == 'Geno':
-            return self.name
-        elif self.dataset.type == 'Publish':
-            return self.post_publication_abbreviation
-        else:
-            return "unnamed"
-
     @property
     def description_fmt(self):
         '''Return a text formated description'''
@@ -252,29 +149,6 @@ class GeneralTrait(object):
             fmt += (' on the minus strand ')
 
         return fmt
-
-# In ProbeSet, there are maybe several annotations match one sequence
-# so we need use sequence(BlatSeq) as the identification, when we update
-# one annotation, we update the others who match the sequence also.
-#
-# Hongqiang Li, 3/3/2008
-def getSequence(trait, dataset_name):
-    dataset = create_dataset(dataset_name)
-     
-    if dataset.type == 'ProbeSet':
-        results = g.db.execute('''
-                       SELECT
-                               ProbeSet.BlatSeq
-                       FROM
-                               ProbeSet, ProbeSetFreeze, ProbeSetXRef
-                       WHERE
-                               ProbeSet.Id=ProbeSetXRef.ProbeSetId and
-                               ProbeSetFreeze.Id = ProbeSetXRef.ProbSetFreezeId and
-                               ProbeSet.Name = %s
-                               ProbeSetFreeze.Name = %s
-               ''', trait.name, dataset.name).fetchone()
-
-        return results[0]        
         
 def retrieve_sample_data(trait, dataset, samplelist=None):
     if samplelist == None:
@@ -293,18 +167,6 @@ def retrieve_sample_data(trait, dataset, samplelist=None):
             if not samplelist or (samplelist and name in samplelist):
                 trait.data[name] = webqtlCaseData(*item)   #name, value, variance, num_cases)
     return trait
-        
-def convert_location_to_value(chromosome, mb):
-    try:
-        location_value = int(chromosome)*1000 + float(mb)
-    except ValueError:
-        if chromosome.upper() == 'X':
-            location_value = 20*1000 + float(mb)
-        else:
-            location_value = (ord(str(chromosome).upper()[0])*1000 +
-                              float(mb))
-
-    return location_value
 
 @app.route("/trait/get_sample_data")
 def get_sample_data():
@@ -314,13 +176,23 @@ def get_sample_data():
 
     trait_ob = GeneralTrait(name=trait, dataset_name=dataset)
 
-    return json.dumps([trait, {key: value.value for key, value in trait_ob.data.iteritems() }])
-
-    #jsonable_sample_data = {}
-    #for sample in trait_ob.data.iteritems():
-    #    jsonable_sample_data[sample] = trait_ob.data[sample].value
-    #
-    #return jsonable_sample_data
+    trait_dict = {}
+    trait_dict['name'] = trait
+    trait_dict['db'] = dataset
+    trait_dict['type'] = trait_ob.dataset.type
+    trait_dict['group'] = trait_ob.dataset.group.name
+    trait_dict['tissue'] = trait_ob.dataset.tissue
+    trait_dict['species'] = trait_ob.dataset.group.species
+    trait_dict['url'] = url_for('show_trait_page', trait_id = trait, dataset = dataset)
+    trait_dict['description'] = trait_ob.description_display
+    if trait_ob.dataset.type == "ProbeSet":
+        trait_dict['symbol'] = trait_ob.symbol
+        trait_dict['location'] = trait_ob.location_repr
+    elif trait_ob.dataset.type == "Publish":
+        trait_dict['pubmed_link'] = trait_ob.pubmed_link
+        trait_dict['pubmed_text'] = trait_ob.pubmed_text
+
+    return json.dumps([trait_dict, {key: value.value for key, value in trait_ob.data.iteritems() }])
     
 def jsonable(trait):
     """Return a dict suitable for using as json
@@ -542,38 +414,7 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False):
             if trait.pubmed_id:
                 trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % trait.pubmed_id
 
-
-        trait.homologeneid = None
         if dataset.type == 'ProbeSet' and dataset.group:
-            if trait.geneid:
-                #XZ, 05/26/2010: From time to time, this query get error message because some geneid values in database are not number.
-                #XZ: So I have to test if geneid is number before execute the query.
-                #XZ: The geneid values in database should be cleaned up.
-                #try:
-                #    float(self.geneid)
-                #    geneidIsNumber = True
-                #except ValueError:
-                #    geneidIsNumber = False
-                #if geneidIsNumber:
-                query = """
-                        SELECT
-                                HomologeneId
-                        FROM
-                                Homologene, Species, InbredSet
-                        WHERE
-                                Homologene.GeneId ='%s' AND
-                                InbredSet.Name = '%s' AND
-                                InbredSet.SpeciesId = Species.Id AND
-                                Species.TaxonomyId = Homologene.TaxonomyId
-                        """ % (escape(str(trait.geneid)), escape(dataset.group.name))
-                logger.sql(query)
-                result = g.db.execute(query).fetchone()
-                #else:
-                #    result = None
-
-                if result:
-                    trait.homologeneid = result[0]
-
             description_string = unicode(str(trait.description).strip(codecs.BOM_UTF8), 'utf-8')
             target_string = unicode(str(trait.probe_target_description).strip(codecs.BOM_UTF8), 'utf-8')
 
@@ -589,46 +430,19 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False):
             # Save it for the jinja2 template
             trait.description_display = description_display
 
-            #XZ: trait_location_value is used for sorting
             trait.location_repr = 'N/A'
-            trait.location_value = 1000000
-
             if trait.chr and trait.mb:
-                #Checks if the chromosome number can be cast to an int (i.e. isn't "X" or "Y")
-                #This is so we can convert the location to a number used for sorting
-                trait_location_value = convert_location_to_value(trait.chr, trait.mb)
-                 #try:
-                #    trait_location_value = int(self.chr)*1000 + self.mb
-                #except ValueError:
-                #    if self.chr.upper() == 'X':
-                #        trait_location_value = 20*1000 + self.mb
-                #    else:
-                #        trait_location_value = (ord(str(self.chr).upper()[0])*1000 +
-                #                               self.mb)
-
-                #ZS: Put this in function currently called "convert_location_to_value"
                 trait.location_repr = 'Chr%s: %.6f' % (trait.chr, float(trait.mb))
-                trait.location_value = trait_location_value
 
         elif dataset.type == "Geno":
             trait.location_repr = 'N/A'
-            trait.location_value = 1000000
-
             if trait.chr and trait.mb:
-                #Checks if the chromosome number can be cast to an int (i.e. isn't "X" or "Y")
-                #This is so we can convert the location to a number used for sorting
-                trait_location_value = convert_location_to_value(trait.chr, trait.mb)
-
-                #ZS: Put this in function currently called "convert_location_to_value"
                 trait.location_repr = 'Chr%s: %.6f' % (trait.chr, float(trait.mb))
-                trait.location_value = trait_location_value
 
         if get_qtl_info:
             #LRS and its location
             trait.LRS_score_repr = "N/A"
-            trait.LRS_score_value = 0
             trait.LRS_location_repr = "N/A"
-            trait.LRS_location_value = 1000000
             if dataset.type == 'ProbeSet' and not trait.cellid:
                 query = """
                         SELECT
@@ -699,19 +513,9 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False):
                     trait.locus = trait.lrs = trait.additive = ""
 
             if (dataset.type == 'Publish' or dataset.type == "ProbeSet") and trait.locus_chr != "" and trait.locus_mb != "":
-                #XZ: LRS_location_value is used for sorting
-                try:
-                    LRS_location_value = int(trait.locus_chr)*1000 + float(trait.locus_mb)
-                except:
-                    if trait.locus_chr.upper() == 'X':
-                        LRS_location_value = 20*1000 + float(trait.locus_mb)
-                    else:
-                        LRS_location_value = ord(str(trait.locus_chr).upper()[0])*1000 + float(trait.locus_mb)
-
                 trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % (trait.locus_chr, float(trait.locus_mb))
                 if trait.lrs != "":
                     trait.LRS_score_repr = LRS_score_repr = '%3.1f' % trait.lrs
-                    trait.LRS_score_value = LRS_score_value = trait.lrs
     else:
         raise KeyError, `trait.name`+' information is not found in the database.'