Rewrote code related to getting the tissue correlation column to display

Created new files for mrna assay tissue data and commonly used
db query related functions

2 files changed, 140 insertions, 94 deletions

diff --git a/wqflask/wqflask/correlation/correlationFunction.py b/wqflask/wqflask/correlation/correlation_functions.py
index 7d4b58a9..56f66810 100644
--- a/wqflask/wqflask/correlation/correlationFunction.py
+++ b/wqflask/wqflask/correlation/correlation_functions.py
@@ -24,6 +24,7 @@
 #
 # Last updated by NL 2011/03/23
 
+from __future__ import absolute_import, print_function, division
 
 import math
 #import rpy2.robjects
@@ -31,10 +32,11 @@ import pp
 import string
 
 from utility import webqtlUtil
+from base.mrna_assay_tissue_data import MrnaAssayTissueData
 from base.trait import GeneralTrait
 from dbFunction import webqtlDatabaseFunction
 
-
+from flask import Flask, g
 
 #XZ: The input 'controls' is String. It contains the full name of control traits.
 #XZ: The input variable 'strainlst' is List. It contains the strain names of primary trait.
@@ -676,7 +678,7 @@ def batchCalTissueCorr(primaryTraitValue=[], SymbolValueDict={}, method='pearson
 # getGeneSymbolTissueValueDict to build dict to get CorrPvArray
 #Note: If there are multiple probesets for one gene, select the one with highest mean.
 ###########################################################################
-def getTissueProbeSetXRefInfo(cursor=None,GeneNameLst=[],TissueProbeSetFreezeId=0):
+def getTissueProbeSetXRefInfo(GeneNameLst=[],TissueProbeSetFreezeId=0):
     Symbols =""
     symbolList =[]
     geneIdDict ={}
@@ -720,7 +722,6 @@ def getTissueProbeSetXRefInfo(cursor=None,GeneNameLst=[],TissueProbeSetFreezeId=
                 '''% (TissueProbeSetFreezeId,Symbols)
 
     try:
-
         cursor.execute(query)
         results =cursor.fetchall()
         resultCount = len(results)
@@ -755,28 +756,43 @@ def getTissueProbeSetXRefInfo(cursor=None,GeneNameLst=[],TissueProbeSetFreezeId=
 #function: get one dictionary whose key is gene symbol and value is tissue expression data (list type).
 #Attention! All keys are lower case!
 ###########################################################################
-def getSymbolValuePairDict(cursor=None,symbolList=None,dataIdDict={}):
-    symbolList = map(string.lower, symbolList)
-    symbolValuepairDict={}
-    valueList=[]
-
-    for key in symbolList:
-        if dataIdDict.has_key(key):
-            DataId = dataIdDict[key]
-
-            valueQuery = "select value from TissueProbeSetData where Id=%s" % DataId
-            try :
-                cursor.execute(valueQuery)
-                valueResults = cursor.fetchall()
-                for item in valueResults:
-                    item =item[0]
-                    valueList.append(item)
-                symbolValuepairDict[key] = valueList
-                valueList=[]
-            except:
-                symbolValuepairDict[key] = None
-
-    return symbolValuepairDict
+def get_symbol_value_pairs(tissue_data):
+    
+    id_list = [tissue_data[symbol.lower()].data_id for item in tissue_data]
+
+    symbol_value_pairs = {}
+    value_list=[]
+
+    query = """SELECT value, id
+               FROM TissueProbeSetData
+               WHERE Id IN {}""".format(create_in_clause(id_list))
+
+    try :
+        results = g.db.execute(query).fetchall()
+        for result in results:
+            value_list.append(result.value)
+        symbol_value_pairs[symbol] = value_list
+    except:
+        symbol_value_pairs[symbol] = None
+
+    #for symbol in symbol_list:
+    #    if tissue_data.has_key(symbol):
+    #        data_id = tissue_data[symbol].data_id
+    #
+    #        query = """select value, id
+    #                   from TissueProbeSetData
+    #                   where Id={}""".format(escape(data_id))
+    #        try :
+    #            results = g.db.execute(query).fetchall()
+    #            for item in results:
+    #                item = item[0]
+    #                value_list.append(item)
+    #            symbol_value_pairs[symbol] = value_list
+    #            value_list=[]
+    #        except:
+    #            symbol_value_pairs[symbol] = None
+
+    return symbol_value_pairs
 
 
 ########################################################################################################
@@ -788,36 +804,51 @@ def getSymbolValuePairDict(cursor=None,symbolList=None,dataIdDict={}):
 #          then call getSymbolValuePairDict function and merge the results.
 ########################################################################################################
 
-def getGeneSymbolTissueValueDict(cursor=None,symbolList=None,dataIdDict={}):
-    limitNum=1000
-    count = len(symbolList)
-
-    SymbolValuePairDict = {}
-
-    if count !=0 and count <=limitNum:
-        SymbolValuePairDict = getSymbolValuePairDict(cursor=cursor,symbolList=symbolList,dataIdDict=dataIdDict)
-
-    elif count >limitNum:
-        SymbolValuePairDict={}
-        n = count/limitNum
-        start =0
-        stop =0
-
-        for i in range(n):
-            stop =limitNum*(i+1)
-            gList1 = symbolList[start:stop]
-            PairDict1 = getSymbolValuePairDict(cursor=cursor,symbolList=gList1,dataIdDict=dataIdDict)
-            start =limitNum*(i+1)
-
-            SymbolValuePairDict.update(PairDict1)
-
-        if stop < count:
-            stop = count
-            gList2 = symbolList[start:stop]
-            PairDict2 = getSymbolValuePairDict(cursor=cursor,symbolList=gList2,dataIdDict=dataIdDict)
-            SymbolValuePairDict.update(PairDict2)
-
-    return SymbolValuePairDict
+def get_trait_symbol_and_tissue_values(symbol_list=None):
+    SymbolValuePairDict={}
+    
+    tissue_data = MrnaAssayTissueData(gene_symbols=symbol_list)
+
+    #symbolList,
+    #geneIdDict,
+    #dataIdDict,
+    #ChrDict,
+    #MbDict,
+    #descDict,
+    #pTargetDescDict = getTissueProbeSetXRefInfo(
+    #                    GeneNameLst=GeneNameLst,TissueProbeSetFreezeId=TissueProbeSetFreezeId)
+    
+    if len(tissue_data.gene_symbols):
+        return get_symbol_value_pairs(tissue_data)
+        
+    #limit_num=1000
+    #count = len(symbol_list)
+    #
+    #symbol_value_pairs = {}
+    #
+    #if count !=0 and count <= limit_num:
+    #    symbol_value_pairs = getSymbolValuePairDict(cursor=cursor,symbolList=symbol_list,dataIdDict=dataIdDict)
+    #
+    #elif count > limit_num:
+    #    n = count/limit_num
+    #    start = 0
+    #    stop = 0
+    #
+    #    for i in range(n):
+    #        stop =limit_num*(i+1)
+    #        gList1 = symbolList[start:stop]
+    #        PairDict1 = getSymbolValuePairDict(cursor=cursor,symbolList=gList1,dataIdDict=dataIdDict)
+    #        start =limit_num*(i+1)
+    #
+    #        SymbolValuePairDict.update(PairDict1)
+    #
+    #    if stop < count:
+    #        stop = count
+    #        gList2 = symbolList[start:stop]
+    #        PairDict2 = getSymbolValuePairDict(cursor=cursor,symbolList=gList2,dataIdDict=dataIdDict)
+    #        SymbolValuePairDict.update(PairDict2)
+    #
+    #return SymbolValuePairDict
 
 ########################################################################################################
 #input: cursor, GeneNameLst (list), TissueProbeSetFreezeId(int)
@@ -827,12 +858,17 @@ def getGeneSymbolTissueValueDict(cursor=None,symbolList=None,dataIdDict={}):
 #          for CorrelationPage.py
 ########################################################################################################
 
-def getGeneSymbolTissueValueDictForTrait(cursor=None,GeneNameLst=[],TissueProbeSetFreezeId=0):
-    SymbolValuePairDict={}
-    symbolList,geneIdDict,dataIdDict,ChrDict,MbDict,descDict,pTargetDescDict = getTissueProbeSetXRefInfo(cursor=cursor,GeneNameLst=GeneNameLst,TissueProbeSetFreezeId=TissueProbeSetFreezeId)
-    if symbolList:
-        SymbolValuePairDict = getGeneSymbolTissueValueDict(cursor=cursor,symbolList=symbolList,dataIdDict=dataIdDict)
-    return SymbolValuePairDict
+#def get_trait_symbol_and_tissue_values(cursor=None,GeneNameLst=[],TissueProbeSetFreezeId=0):
+#    SymbolValuePairDict={}
+#    
+#    symbolList,geneIdDict,dataIdDict,ChrDict,MbDict,descDict,pTargetDescDict = getTissueProbeSetXRefInfo(
+#        cursor=cursor,GeneNameLst=GeneNameLst,TissueProbeSetFreezeId=TissueProbeSetFreezeId)
+#    
+#    if symbolList:
+#        SymbolValuePairDict = get_gene_symbol_and_tissue_values(symbolList=symbolList,
+#                                                                dataIdDict=dataIdDict)
+#    
+#    return SymbolValuePairDict
 
 ########################################################################################################
 #Input: cursor(cursor): MySQL connnection cursor;
diff --git a/wqflask/wqflask/correlation/show_corr_results.py b/wqflask/wqflask/correlation/show_corr_results.py
index 4a9aea73..d1a7ff15 100644
--- a/wqflask/wqflask/correlation/show_corr_results.py
+++ b/wqflask/wqflask/correlation/show_corr_results.py
@@ -56,6 +56,8 @@ from MySQLdb import escape_string as escape
 
 from pprint import pformat as pf
 
+from flask import Flask, g
+
 METHOD_SAMPLE_PEARSON = "1"
 METHOD_SAMPLE_RANK = "2"
 METHOD_LIT = "3"
@@ -161,6 +163,11 @@ class CorrelationResults(object):
                 trait_object.sample_r = self.correlation_data[trait][0]
                 trait_object.sample_p = self.correlation_data[trait][1]
                 trait_object.num_overlap = self.correlation_data[trait][2]
+                
+                # Set some sane defaults
+                trait_object.tissue_corr = None
+                trait_object.tissue_pvalue = None
+
                 self.correlation_results.append(trait_object)
                 
                 #self.correlation_data_slice[trait] = self.correlation_data[trait]
@@ -593,7 +600,7 @@ class CorrelationResults(object):
         return litCorrDict
 
 
-    def fetch_tissue_correlations(self, method=""):
+    def fetch_tissue_correlations(self):
         """Comments Possibly Out of Date!!!!!
         
         
@@ -939,39 +946,42 @@ class CorrelationResults(object):
         return (symbolCorrDict, symbolPvalueDict)
 
 
+    def do_tissue_correlation_by_list(self, tissue_dataset_id):
 
-    #XZ, 10/13/2010
-    def getTissueCorrelationByList(self, primaryTraitSymbol=None, traitList=None, TissueProbeSetFreezeId=None, method=None):
-
-        primaryTraitSymbolValueDict = correlationFunction.getGeneSymbolTissueValueDictForTrait(cursor=self.cursor, GeneNameLst=[primaryTraitSymbol], TissueProbeSetFreezeId=TISSUE_MOUSE_DB)
-
-        if primaryTraitSymbol.lower() in primaryTraitSymbolValueDict:
-            primaryTraitValue = primaryTraitSymbolValueDict[primaryTraitSymbol.lower()]
-
-            geneSymbolList = []
-
-            for thisTrait in traitList:
-                if hasattr(thisTrait, 'symbol'):
-                    geneSymbolList.append(thisTrait.symbol)
+        trait_symbol_and_values = correlation_function.get_trait_symbol_and_tissue_values(
+            gene_name_list = [self.this_trait.symbol])
 
-            SymbolValueDict = correlationFunction.getGeneSymbolTissueValueDictForTrait(cursor=self.cursor, GeneNameLst=geneSymbolList, TissueProbeSetFreezeId=TISSUE_MOUSE_DB)
-
-            for thisTrait in traitList:
-                if hasattr(thisTrait, 'symbol') and thisTrait.symbol and thisTrait.symbol.lower() in SymbolValueDict:
-                    oneTraitValue = SymbolValueDict[thisTrait.symbol.lower()]
-                    if method in ["2","5"]:
-                        result = correlationFunction.calZeroOrderCorrForTiss( primaryTraitValue, oneTraitValue, method='spearman' )
-                    else:
-                        result = correlationFunction.calZeroOrderCorrForTiss( primaryTraitValue, oneTraitValue)
-                    thisTrait.tissueCorr = result[0]
-                    thisTrait.tissuePValue = result[2]
-                else:
-                    thisTrait.tissueCorr = None
-                    thisTrait.tissuePValue = None
-        else:
-            for thisTrait in traitList:
-                thisTrait.tissueCorr = None
-                thisTrait.tissuePValue = None
+        if self.this_trait.symbol.lower() in trait_symbol_and_values:
+            primary_trait_value = trait_symbol_and_values[self.this_trait_symbol.lower()]
+            
+            #gene_symbol_list = []
+            #
+            #for trait in self.correlation_results:
+            #    if hasattr(trait, 'symbol'):
+            #        gene_symbol_list.append(trait.symbol)
+            
+            gene_symbol_list = [trait.symbol for trait in self.correlation_results if trait.symbol]
+
+            symbol_value_dict = correlation_function.get_trait_gene_symbol_and_tissue_values(
+                                                    gene_symbol_list=gene_symbol_list)
+
+            for trait in self.correlation_results:
+                if trait.symbol and trait.symbol.lower() in symbol_value_dict:
+                    this_trait_value = symbol_value_dict[trait.symbol.lower()]
+                    
+                    result = correlation_function.calZeroOrderCorrForTiss(primary_trait_value,
+                                                                          this_trait_value,
+                                                                          self.corr_method)
+ 
+                    trait.tissue_corr = result[0]
+                    trait.tissue_pvalue = result[2]
+        #        else:
+        #            trait.tissue_corr = None
+        #            trait.tissue_pvalue = None
+        #else:
+        #    for trait in self.correlation_results:
+        #        trait.tissue_corr = None
+        #        trait.tissue_pvalue = None
 
-        return traitList
+        #return self.correlation_results