Several interface and computational changes.

1-Moved the normweighmode to the beginning of the program.
In this way, the weight will be applied at the beginning
instead of just the final normalization step. Our tests
show that it improves the results when Max(JET, (PC+CV)2)
is used at the beginning. Most of the changes in this part
are in computePred.R file.
2-When the user provides a mutation list
as a file, it was giving an error. The error in apply(...)
part of normalizePredWithNbSeqsPCSelMult function. The bug
is corrected now and it produces results consistent with
the web server.

3-A small bug in gemme.py was also corrected.
If a mutation list file is given, the model is not 'simple'
and therefore, it will not produce a map of single point
mutations.

gemme.py

@@ -118,7 +118,8 @@ def calcBfactors(pdbfile, outfile, nmodes=None, attenuate="true",
         else:
             return bfactors
 
-def getBfactors(pdbfile, outfile, nmodes=None, attenuate="true", ranksorted="true"):
+def getBfactors(pdbfile, outfile, attenuate="true", \
+                ranksorted="true",  inverted=True):
     """
         Calculate rank normalized dynamic flexibility index. 
     """
@@ -141,10 +142,15 @@ def getBfactors(pdbfile, outfile, nmodes=None, attenuate="true", ranksorted="tru
     #     np.savetxt(outfile, dfi)
 
     if(ranksorted.lower()=='true'):
-        return percentBfactors
+        if(inverted==True):
+            return (1.0 - percentBfactors)
+        else:
+            return percentBfactors
     else:
-        return bfactors
-
+        if(inverted==True):
+            return (np.max(bfactors) - bfactors)
+        else:
+            return bfactors
     
 def attenuateEndPoints(dfi):
     """
@@ -437,15 +443,15 @@ def parse_command_line():
         required=False, default=None)
 
     retMet_args.add_argument('--alphabet', dest='alphabet', type=str, \
-        help="Which alphabet to use. Default is lz-bl.7",
-        required=False, default="lz-bl.7")
+        help="Which alphabet to use. Default is lw-i.7",
+        required=False, default="lw-i.7")
 
     retMet_args.add_argument('-p', '--pdbfile', dest='pdbfile', type=str, \
         help="If a pdb file is provided, it will skip fake pdb file production step and use that file. Default is None",
         required=False, default=None)
     
     retMet_args.add_argument('--normweightmode', dest='normweightmode', type=str, \
-        help="It can be one of these: 'trace', 'cv', 'pc', 'dfi', 'max-trace-pc', 'max-trace-cv', 'max-trace-pc-cv' or 'half-cv+pc'. Default is 'trace'.",
+        help="It can be one of these: 'trace', 'tracemovingaverage', 'cv', 'pc', 'dfi', 'maxtracepc', 'maxtracecv', 'maxtracepccv' or 'halfcvpc'. Default is 'trace'.",
         required=False, default="trace")
 
     args = parser.parse_args()
@@ -467,7 +473,10 @@ def doit(inAli,mutFile,retMet,bFile,fFile,n,N, jetfile, pdbfile, normWeightMode,
     doit is basically the main function in disguise!
     doit(args.input,args.mutations,args.retrievingMethod,args.blastFile,args.fastaFile, args.jetfile)
     """
-    simple = True
+    if(mutFile != ''):
+        simple = False
+    else:
+        simple = True
 
     
     prot,seq,nl=extractQuerySeq(inAli)
@@ -488,27 +497,28 @@ def doit(inAli,mutFile,retMet,bFile,fFile,n,N, jetfile, pdbfile, normWeightMode,
         (normWeightMode != 'pc') and \
         (normWeightMode != 'dfi') and \
         (normWeightMode != 'bfactor') and \
-        (normWeightMode != 'max-trace-pc') and \
-        (normWeightMode != 'max-pc-trace') and \
-        (normWeightMode != 'max-pc-cv') and \
-        (normWeightMode != 'max-cv-pc') and \
-        (normWeightMode != 'max-trace-cv') and \
-        (normWeightMode != 'max-cv-trace') and \
-        (normWeightMode != 'max-trace-dfi') and \
-        (normWeightMode != 'max-trace-bfactor') and \
-        (normWeightMode != 'max-dfi-trace') and \
-        (normWeightMode != 'max-bfactor-trace') and \
-        (normWeightMode != 'max-trace-pc-cv') and \
-        (normWeightMode != 'max-trace-cv-pc') and \
-        (normWeightMode != 'half-pc+cv') and \
-        (normWeightMode != 'half-cv+pc') and \
-        (normWeightMode != 'max-trace-half-cv+pc') and \
-        (normWeightMode != 'max-trace-half-pc+cv')):
-        print("ERROR: normWeightMode can only be 'trace', 'cv', 'pc', 'dfi', bfactor,\n"+\
-              "       'max-trace-pc', 'max-trace-dfi', 'max-pc-cv', "+\
-              "       'max-trace-bfactor'"+\
-              "       'max-trace-cv', 'max-trace-pc-cv'"+\
-              "       'half-cv+pc' or max-trace-half-cv+pc!")
+        (normWeightMode != 'maxtracepc') and \
+        (normWeightMode != 'maxpctrace') and \
+        (normWeightMode != 'maxpccv') and \
+        (normWeightMode != 'maxcvpc') and \
+        (normWeightMode != 'maxtracecv') and \
+        (normWeightMode != 'maxcvtrace') and \
+        (normWeightMode != 'maxtracedfi') and \
+        (normWeightMode != 'maxtracebfactor') and \
+        (normWeightMode != 'maxdfitrace') and \
+        (normWeightMode != 'maxbfactortrace') and \
+        (normWeightMode != 'maxtracepccv') and \
+        (normWeightMode != 'maxtracecvpc') and \
+        (normWeightMode != 'halfpccv') and \
+        (normWeightMode != 'halfcvpc') and \
+        (normWeightMode != 'maxtracehalfcvpc') and \
+        (normWeightMode != 'tracemovingaverage') and \
+        (normWeightMode != 'maxtracehalfpccv')):
+        print("ERROR: normWeightMode can only be 'trace', 'tracemovingaverage', 'cv', 'pc', 'dfi', bfactor,\n"+\
+              "       'maxtracepc', 'maxtracedfi', 'maxpccv', "+\
+              "       'maxtracebfactor'"+\
+              "       'maxtracecv', 'maxtracepccv'"+\
+              "       'halfcvpc' or maxtracehalfcvpc!")
         sys.exit(-1)
 
     structure = parsePDB(prot+".pdb")
@@ -537,7 +547,7 @@ def doit(inAli,mutFile,retMet,bFile,fFile,n,N, jetfile, pdbfile, normWeightMode,
         print("done")
 
     #If a real pdb file is given, calculate dfi for the residues. 
-    if(((normWeightMode=='dfi') or (normWeightMode=='max-trace-dfi') or (normWeightMode=='max-dfi-trace'))):
+    if(((normWeightMode=='dfi') or (normWeightMode=='maxtracedfi') or (normWeightMode=='maxdfitrace'))):
         if (pdbfile == None):
             print("ERROR: There is not any pdb file to calculate DFI.")
             sys.exit(-1)
@@ -553,21 +563,49 @@ def doit(inAli,mutFile,retMet,bFile,fFile,n,N, jetfile, pdbfile, normWeightMode,
 
             print(df)
 
+    #If a real pdb file is given, calculate dfi for the residues. 
+    if(((normWeightMode=='tracemovingaverage'))):
+        print("Calculating trace moving average per residue!")
+        # tma = tracemovingaverage
+        df = pd.read_csv(prot+"_jet.res", delimiter=r"\s+") 
+        print(df.columns)
+        # Get the window of series
+        # of observations of specified window size
+        window_size = 3
+        windows = df['trace'].rolling(window_size)
+        
+        # Create a series of moving
+        # averages of each window
+        moving_averages = windows.mean()
+        tma = moving_averages
+
+        print(df)
+
+        df['tracemovingaverage'] = tma.round(4)
+        #The first two elements are just copied bc they don't exist due to the moving average. 
+        df['tracemovingaverage'].iloc[0] = df['trace'].iloc[0].round(4)
+        df['tracemovingaverage'].iloc[1] = df['trace'].iloc[1].round(4)
+        df.to_csv(prot+"_jet.res", header=True, index=None, sep='\t', mode='w')
 
+        print(df)
     #If a real pdb file is given, calculate or get Bfactors for the residues. 
-    if(((normWeightMode=='bfactor') or (normWeightMode=='max-trace-bfactor') or (normWeightMode=='max-bfactor-trace'))):
-        isCalc = False
+    if(((normWeightMode=='bfactor') or (normWeightMode=='maxtracebfactor') or (normWeightMode=='maxbfactortrace'))):
+        isCalc = True
         if (pdbfile == None):
             print("ERROR: There is not any pdb file to calculate or get Bfactors!")
             sys.exit(-1)
         else:        
             if (isCalc):
                 print("Computing Bfactors from the user-provided pdb file using ANM with all modes.")
-                bfactors = calcBfactors(pdbfile, outfile=None, nmodes=None, \
-                            attenuate="true", ranksorted="true", inverted=False)
+                #Original
+                # bfactors = calcBfactors(pdbfile, outfile=None, nmodes=None, \
+                #           attenuate="true", ranksorted="true", inverted=False)
+                bfactors = calcBfactors(pdbfile, outfile=None, nmodes=10, \
+                            attenuate="false", ranksorted="true", inverted=True)
             else:
                 print("Getting Bfactors from the user-provided pdb file.")
-                bfactors = getBfactors(pdbfile, outfile=None, nmodes=None, attenuate="true", ranksorted="true")
+                bfactors = getBfactors(pdbfile, outfile=None,\
+                            attenuate="true", ranksorted="true", inverted=False)
 
             bfactors = bfactors
             df = pd.read_table(prot+"_jet.res")

pred.R

@@ -358,12 +358,12 @@ computePredSimple<-function(mat, distTrace, wt, thresh){
         if(length(sel)>0){
           sortedDist=sort(distTrace[sel-1])
 
-          if((i==694) & (a=="v")) {
-            print(i)
-            print(sel)
-            print(distTrace[sel-1])
-            print(sortedDist)
-            }
+          # if((i==694) & (a=="v")) {
+          #   print(i)
+          #   print(sel)
+          #   print(distTrace[sel-1])
+          #   print(sortedDist)
+          #   }
           
           }
         if(length(sel)>0){sortedDist=sort(distTrace[sel-1])}