Added cluster persistence plots

-,iarroyof,bisnaga,19.02.2018 13:28,file:///home/iarroyof/.config/libreoffice/4;
\ No newline at end of file
+,iarroyof,bisnaga,20.02.2018 21:25,file:///home/iarroyof/.config/libreoffice/4;
\ No newline at end of file

+
+from pdb import set_trace as st
+import os
+import matplotlib.pyplot as plt
+
+
+def analyze_pair(group_anlized, gropus_dir, rev, eigenvectors):
+    dic_part = {}
+    with open(eigenvectors) as f:
+
+        eigenvector = [c for c in f.readlines() if c.startswith("(" +
+                                            str(group_anlized - 1) + ",")][0]
+    x = eigenvector.split(" ")[1]
+    eigen = {"string": x,
+            "bo": x.split("*")[1].replace('"', ''),
+            "value": float(x.split("*")[0].replace("u'", '').strip())
+            }
+
+    filelist = os.listdir(gropus_dir)
+
+    init_gfile = "gus_originales_" + str(rev).zfill(3) + ".cls"
+
+    with open(gropus_dir + init_gfile, 'r') as f:
+        groups = f.readlines()
+        del groups[-1]
+        tfs = [c for c in groups if c.startswith(str(group_anlized))][0]
+
+        print("The analized group = %s" % tfs.split('\t')[0])
+        print("The corresponding TFs: %s" % tfs.split('\t')[1])
+        print("The corresponding eigenvector: %s" % eigenvector)
+        group = {}
+        ss = tfs.split(",")
+
+        TFs = {[s for s in tf.strip().partition("(")[0].partition("\t")
+            if s.replace('-', '').isalpha()][0]: [float(
+                tf.partition("(")[-1].strip().strip(")"))] for tf in ss}
+
+    partition = []
+    for i in reversed(range(2, rev)):
+    #if i.endswith(".cls"):  # You could also add "and i.startswith('f')
+        name = "gus_originales_" + str(i).zfill(3) + ".cls"
+        with open(gropus_dir + name, 'r') as f:
+            gropus = f.readlines()
+            del gropus[-1]
+
+            for c in gropus:
+                ss = c.split(",")
+                TFss = {[s for s in tf.strip().partition("(")[0].partition("\t")
+                     if s.replace('-', '').isalpha()][0]: [float(
+                             tf.partition("(")[-1].strip().strip(")"))]
+                                                                for tf in ss}
+
+                if set((t for t in list(TFs.keys()))) <= set(TFss):  # verify if TFs
+                    not_in = set(TFs.keys()) ^ set(list(TFss.keys()))  # are in the
+                    for k in list(TFs.keys()):                    # current cluster
+                        TFs[k] += TFss[k]
+
+                    if not_in:
+                        for n in not_in:
+                            TFs[n] = TFss[n]
+
+        partition.append((i, list(TFs.keys()), len(list(TFs.keys()))))
+
+    partition = zip(*partition)
+    dic_part["part"] = partition[0]
+    dic_part["tfs"] = partition[1]
+    dic_part["n_tfs"] = partition[2]
+
+    return dic_part
+
+
+def get_cmap(n, name='hsv'):
+    '''Returns a function that maps each index in 0, 1, ..., n-1 to a distinct
+    RGB color; the keyword argument name must be a standard mpl colormap name.'''
+    return plt.cm.get_cmap(name, n)
+
+rev = 120
+
+eigenvectors = "one-by-one/eigen_vectors/eigenBOs_120-eigens.txt"
+gropus_dir = "one-by-one/groups/"
+
+# Candidatos que inician con 2 o 3 TFs desde la particion de 120 grupos
+groups_analyzed = [97, 80, 74, 68, 63, 53, 52, 49, 47, 44, 43, 40, 39, 38, 37,
+                    36, 34, 32, 31, 30, 29, 27, 26, 24, 23, 21, 20, 19, 18, 15,
+                    13, 12, 9]
+cmap = get_cmap(len(groups_analyzed))
+
+plots = []
+
+for g in groups_analyzed:
+    partition = analyze_pair(group_anlized=g, gropus_dir=gropus_dir,
+                                        rev=rev, eigenvectors=eigenvectors)
+    plots.append(partition)
+
+fig = plt.figure()
+ax1 = fig.add_subplot(111)
+
+for i, p in enumerate(plots):
+    if " ".join(p['tfs'][0]) == "Zur ZntR":
+        width = 4
+        mark = "D"
+    else:
+        width = 2
+        mark = ""
+    ax1.plot(p['part'], p['n_tfs'], c=cmap(i), linewidth=width, marker=mark,
+                                                label=" ".join(p['tfs'][0]))
+
+#ax1.scatter(x[40:],y[40:], s=10, c='r', marker="o", label='second')
+plt.legend(loc='upper right')
+plt.title("TF pair persistence through partitions of model resolution")
+plt.xlabel("Partition")
+plt.ylabel("Number of TFs")
+plt.show()
+
+#print(TFs.keys())
+#print(TFss.keys())
+#partition = zip(*partition)
+#print(partition)