added window scanning

models/test_pickle.py

@@ -3,7 +3,7 @@ import networkx as nx
 from matplotlib import pyplot as plt
 
 #g = pickle.load(open("all_graphs_pickled/" + "1FFK" + ".nxpickled", "rb"))
-PDBid = "1EBR.A"
+PDBid = "1Q7Y.A"
 #print("PDB")
 #print(PDBid)
 PDB, chain = PDBid.split(".")

src/carnaval_to_rmdetect.py

@@ -63,6 +63,65 @@ def get_consensus_sequence(seqs):
         consensus.append(best)
     return consensus
 
+def get_PDB_sequence(PDBid):
+    print("PDB")
+    print(PDBid)
+    PDB, chain = PDBid.split(".")
+    #print(PDB)
+    #print("../all_graphs_pickled/" + PDB + ".nxpickled")
+    try:
+        g = pickle.load(open("../models/all_graphs_pickled/" + PDB + ".nxpickled", "rb"))
+    except FileNotFoundError:
+        print("PDB FILE NOT FOUND")
+        return ("",0)
+    seq=""
+    nodes= []
+    for node in g.nodes(data=True):
+        #print(node)
+        #print(node[0][0],chain)
+        if node[0][0]==chain:
+            nodecode = (node[0][1])
+            nodes.append((int(nodecode),node[1]["nt"]))
+    sortednodes = sorted(list(nodes))
+    nuc_by_node={}
+    missing_nuc = False
+    #print("NODES")
+    numericals = [x[0] for x in sortednodes]
+    decalage = 0
+    if 1 not in numericals:
+        decalage=decalage+1
+        sortednodes.append((1,"N"))
+        sortednodes = sorted(sortednodes)
+        #missing_nuc=True
+        #decalage = decalage +1
+        newnodes = []
+        #for i in sortednodes:
+        #    newnodes.append((i[0],i[1]))
+        #sortednodes = sorted(list(newnodes))
+        numericals = [x[0] for x in sortednodes]
+        #print("MISSING 1", PDBid)
+        #print(numericals)
+        #
+        #sortednodes=sorted(sortednodes)
+        #numericals = [x[0]-1 for x in sortednodes]
+        #numericals.insert(0,0)
+    #else:
+        #print("NOT MISSING", PDBid)
+    for i in sortednodes:
+        nuc_by_node[i[0]]=i[1]
+    #print(sortednodes)
+    for i in range(1,int(sortednodes[-1][0])+1):
+        if i not in numericals:
+            "NOT IN NODES"
+            seq = seq+"-"
+        else:
+            seq = seq + nuc_by_node[i]
+    ss = g.graph['ss']
+    #print(seq)
+    #print("MISSING_NUC",PDBid,missing_nuc)
+    return(seq,decalage)
+
+
 with open("seq.fasta", "a") as f:
     for i in seqs:
         f.write(">seq" + str(i) + "\n")
@@ -78,6 +137,14 @@ with open("seq.fasta", "a") as f:
 print(graphs[module][2].nodes(data=True))
 ss= get_ss_from_graph(graphs[module][0])
 print(ss)
-print(PDBs[module][2])
-print(PDB_positions[module][2])
+print(PDBs[module][86])
+print(PDB_positions[module][86])
+
+
+
 
+#for pos,i in enumerate(PDBs[module][1:]):
+#    #print(i)
+#    seq,dec = get_PDB_sequence(".".join((i.split(".")[:2])))
+#    print(pos, len(seq))
+#TEST : 86

src/module_1.stk

@@ -876,6 +876,4 @@ seq290 CUUCGG
 seq291 CUUCGG
 #=GS seq291 AC seq291
 #=GS seq291 DE seq291
-REF_SEQ CUUCGG
-#=GC SS_cons    <....>
 //

src/parse_sequences.py

@@ -69,13 +69,19 @@ def run_fasta(input, modules_to_parse,dataset,ss = "",arguments={}):
             input = str(input).replace("T","U")
         if len(input)<=300:
             maxs = run_BP(input, ss, modules_to_parse, dataset, "NONE",m,n,sm,mc,k)
+            print("FINAL RESULTS:")
             print(maxs)
         else:
-            maxs = run_BP(input[len(seq)-300:], ss, modules_to_parse, dataset, "NONE",m,n,sm,mc,k)
-            print(maxs)
-
-
-
+            all_maxes=[]
+            index = 0
+            while index+100<len(seq):
+                maxs = run_BP(input[index:index+100], ss, modules_to_parse, dataset, "NONE",m,n,sm,mc,k)
+                #print(maxs)
+                all_maxes.append(maxs)
+            maxs = run_BP(input[index:], ss, modules_to_parse, dataset, "NONE",m,n,sm,mc,k)
+            all_maxes.append(maxs)
+            print("FINAL RESULTS:")
+            print(all_maxes)
 
 if __name__ == "__main__":
     arguments = {}
@@ -124,7 +130,7 @@ if __name__ == "__main__":
 
     graphs = pickle.load(open("../models/"+dataset+"_one_of_each_graph.cPickle", "rb"))
 
-    run_fasta(seq,range(len(graphs)),dataset,ss,arguments)
+    #run_fasta(seq,range(len(graphs)),dataset,ss,arguments)
     #run_fasta(seq,[0,1],dataset,ss,arguments)
-
+    run_fasta(seq,2,dataset,ss,arguments)