.

Aggregation/simi.py

@@ -0,0 +1,58 @@
+#!/bin/env python
+
+# usage: simi.py (similar|exact) inputfile 
+# input file format:
+# smiles identifier
+#
+# Contributing authors:
+#   Teague Sterling, Pascal Wassam, Gurgen Turmanian and John Irwin, 2010-2013
+#
+
+import xmlrpclib, sys, pprint
+
+
+def checkAggregators(smi):
+    result = server.similarity.getSimilars(smi)
+    try:
+        similarCount = len(result['aggregators'])
+
+        for r in result['aggregators']:
+            if r['similarity'] == 1.0:
+                exactMatch = True
+                exactmatchref =  r['reference']
+        return (similarCount, exactMatch, exactmatchref)
+
+    except:
+        pass
+
+
+lines = file(sys.argv[1]).readlines()
+lookups = {}
+myfile = sys.argv[2]
+for l in lines:
+    t = l.split(',')
+    smi = t[1]
+    id = t[0]
+    activity = t[2]
+    if len(id) == 0: id = None
+    lookups[id] = smi
+
+
+server_url = "http://advisor.bkslab.org:8080/aggregate_lookup/xml-rpc/"
+server = xmlrpclib.ServerProxy(server_url);
+similarCount = 0
+exactMatch = False
+outfile = open(myfile, 'w')
+outfile.write('ID, Activity, No. of matches,Exact match, Ref for exact match')
+for (id, smi) in lookups.iteritems():
+    try:
+        checkResult = checkAggregators(smi)
+        try:
+            outfile.write(id+","+activity+","+",".join(str(x) for x in checkResult)+"\n")
+            print id, checkResult
+        except:
+            outfile.write(id+ ","+ activity +","+ "None"+"\n")
+            print id, checkResult
+    except:
+        pass
+outfile.close()

accuAnalysis/combine.py

@@ -0,0 +1,27 @@
+from dipan_utilities import utilities
+import getUnique as gt
+import exportOverlaps as ex
+
+overlaps13 = ex.exportOverlaps(0, 2)
+overlaps23 = ex.exportOverlaps(1, 2)
+overlaps12 = ex.exportOverlaps(0, 1)
+desktopDir = '/Users/dghosh/Desktop/'
+
+#Accounting for all the molecules showing class changing
+switchersForset1 = utilities.extractList(overlaps13[0], gt.set3dict)
+switchersForset2 = utilities.extractList(overlaps23[0], gt.set3dict)
+switchersForset2and1 = utilities.extractList(overlaps12[0], gt.set2dict)
+switchersForset2and1From1 = utilities.extractList(overlaps12[0], gt.set1dict)
+
+print len(switchersForset2and1From1)
+
+set3list = utilities.extractList(gt.set3idSet, gt.set3dict)
+uniqFromSet1 = utilities.extractList(gt.uniq1to3, gt.set1dict)
+uniqFromSet2 = utilities.extractList(gt.uniq2to3, gt.set2dict)
+
+combined = set.union(set(set3list), set(uniqFromSet1), set(uniqFromSet2))
+
+
+combinedExcludeSwitching =(set(switchersForset1)|set(switchersForset2)|set(switchersForset2and1)|set(switchersForset2and1From1))
+print len(combinedExcludeSwitching)
+utilities.listToFile(desktopDir+'toExclude.txt', combinedExcludeSwitching)

accuAnalysis/determine_expt_inacc.py

@@ -44,8 +44,10 @@ def calculateBalAccforSets(i, j):
     activeinCommon = float(intersectionByIndex(i, j, givefullset=False))
     totalinCommon = float(intersectionByIndex(i, j, givefullset=True))
     inactiveinCommon = totalinCommon - activeinCommon
-    balAcc = 0.5 * ((switched[0] / activeinCommon) + switched[1] / inactiveinCommon)
-    return balAcc
+    TP = activeinCommon - switched[0]
+    TN = inactiveinCommon - switched[1]
+    balAcc = 0.5 * ((TP / (TP + switched[0])) + (TN/ (TN + switched[1])))
+    print balAcc
 
 
 if __name__ == '__main__':

accuAnalysis/exportoverlap_implement.py

@@ -2,8 +2,9 @@
 import loadfiles as ld
 from dipan_utilities import utilities
 
-overlaps = ex.exportOverlaps(1, 2)
-switchFile = open(ld.desktopDir + 'set2to3SwitchedMID.csv', 'w')
-notSwitchFile = open(ld.desktopDir + 'set2to3NotSwitchedMID.csv', 'w')
-utilities.pullFromDictAndWriteToFIle(ld.set2dict, overlaps[0], switchFile)
-utilities.pullFromDictAndWriteToFIle(ld.set2dict, overlaps[1], notSwitchFile)
+overlaps12 = ex.exportOverlaps(0, 1)
+switchFile = open(ld.desktopDir + 'set1to2Switched.csv', 'w')
+NotswitchFile = open(ld.desktopDir + 'set1to2NotSwitched.csv', 'w')
+utilities.pullFromDictAndWriteToFIle(ld.set2dict, overlaps12[0], switchFile)
+utilities.pullFromDictAndWriteToFIle(ld.set2dict, overlaps12[1], NotswitchFile)
+

accuAnalysis/fishOut.py

@@ -0,0 +1,13 @@
+import loadfiles as ld
+from dipan_utilities import utilities
+
+with open('/Users/dghosh/Desktop/set3_aggregators') as f:
+    set3Aggre = f.read().splitlines()
+with open('/Users/dghosh/Desktop/set3-nonAggregators') as f:
+    set3NonAggre = f.read().splitlines()
+
+switchFile = open(ld.desktopDir + 'set3AggreRID.csv', 'w')
+notSwitchFile = open(ld.desktopDir + 'set3nonAggreRID.csv', 'w')
+
+utilities.pullFromDictAndWriteToFIle(ld.set3dict, set3Aggre, switchFile)
+utilities.pullFromDictAndWriteToFIle(ld.set3dict, set3NonAggre, notSwitchFile)

accuAnalysis/getUnique.py

@@ -13,6 +13,8 @@
 uniq1to3 = set1idSet - set3idSet
 uniq2to1 = set2idSet - set1idSet
 uniq3to1 = set3idSet - set1idSet
+uniq2to3 = set2idSet - set3idSet
+uniq3to2 = set3idSet - set2idSet
 
 def extractList(idlist, mydict):
     mylist = []

dipan_utilities/utilities.py

@@ -4,7 +4,7 @@ def csvToDict(filename):
     reader = csv.reader(open(filename, 'r'))
     dict = {}
     for row in reader:
-       _, v, k = row
+       v, _, k = row
        dict[k] = v
     return dict
 
@@ -18,4 +18,13 @@ def pullFromDictAndWriteToFIle(yourDict, yourSet, yourFile):
         try:
             yourFile.write(yourDict[id] + '\n')
         except:
-            pass
+            pass
+
+def extractList(idlist, mydict):
+    mylist = []
+    for id in idlist:
+        try:
+            mylist.append(mydict[id])
+        except KeyError:
+            print id + " Key does not exist here!"
+    return mylist

learning/fizzbuzz.py

@@ -0,0 +1,10 @@
+for i in range(100):
+    if i%3 == 0:
+            if i%5==0:
+                print 'fizzbuzz'
+            else:
+                print 'fizz'
+    elif i%5 == 0:
+        print 'buzz'
+    else:
+        print i

writegridCode.py

@@ -0,0 +1,11 @@
+'''
+Created on 13 Mar 2017
+
+@author: dghosh
+'''
+import os
+dirName = "/Users/dghosh/Creative Cloud Files/portfolio_site/rome/Rome-web/day1"
+os.chdir(dirName)
+for filename in os.listdir("."):
+    basename = filename[:-8]
+    print '<a href="./'+dirName.split("/")[-1]+'/'+basename+'-web.jpg" data-toggle="lightbox" data-gallery="christmasmarket"><figure class="photothumbnail"><img src="./'+dirName.split("/")[-1]+'/thumb/'+basename+'-web-thumb.jpg" alt="'+filename+'" class="center-block"></figure></a>'