small fixes

pipelines/gd_tendency.py

@@ -14,10 +14,6 @@ target_category = sys.argv[2]
 target_bed = sys.argv[3]
 marks_dataset = sys.argv[4]
 
-# cell_type = "H1-hESC"
-# target_category = "nre"
-# target_bed = "hg38_NRE_filt5"
-# marks_dataset = "Encode"
 
 # Statistical test
 Test = 'mannwhitneyu'
@@ -174,6 +170,13 @@ for m in marks:
 dfMarks = pd.DataFrame.from_dict(dicMarks, orient = 'index')
 dfMarks.columns = columns
 
+# Order Marks
+orderMarks = []
+for num, what in enumerate(['Acethylation', 'Methylation'], 1):
+    orderMarks.extend(sorted([m for m in dicMarks.keys() if what[:2].lower() in m]))
+
+dfMarks = dfMarks.reindex(orderMarks)
+
 
 # Save statistic tab
 tab_result = os.path.join(target_dir, marks_dataset + "_results", 'statistics', 'test')
@@ -200,7 +203,7 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
         y = np.mean(NormData[m][order[-1]])
         if target_category == 'genes':
             y2 = 0.05 + 0.8 * n / len(these_marks)
-            plt.ylim(0, 0.8)
+            plt.ylim(0, 0.9)
         else:
             y2 = 0.05 + 7 * n / len(these_marks)
             #plt.ylim(0, 6)
@@ -215,9 +218,11 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
     #plt.title(what)
     #plt.xlabel(what,fontsize=12)
     if target_category == "genes":
-        plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=12)
+        plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=12, labelpad=10)
     else:
-        plt.ylabel('Normalized histone marks per length of RIs',fontsize=12)
+        plt.ylabel('Normalized histone marks per length of RIs',fontsize=12, labelpad=10)
+    plt.xlabel('Division of compartments in the nucleus',fontsize=12, labelpad=10)
+    axe.xaxis.set_label_coords(0.32,-0.2)
     #fig.suptitle('test title', fontsize=20)
     axe.spines['right'].set_visible(False)
     axe.spines['top'].set_visible(False)
@@ -229,7 +234,7 @@ if not os.path.exists(plot_path):
     os.makedirs(plot_path)
 
 print("Generating general plot")
-plt.savefig(os.path.join(plot_path, 'normalized_marks.png'))
+plt.savefig(os.path.join(plot_path, 'normalized_marks.png'), dpi=300)
 plt.close()
 
 
@@ -259,9 +264,10 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
         plt.plot([0, len(order) - 1], [0] * 2, 'k-')
         plt.title(str(what + "("+ m +")"))
         if target_category == "genes":
-            plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=12)
+            plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=14, labelpad=10)
         else:
-            plt.ylabel('Normalized histone marks per length of RIs',fontsize=12)
+            plt.ylabel('Normalized histone marks per length of RIs',fontsize=14 , labelpad=10)
+        plt.xlabel('Division of compartments in the nucleus',fontsize=14, labelpad=10)
         axe.spines['right'].set_visible(False)
         axe.spines['top'].set_visible(False)
         axe.spines['bottom'].set_visible(False)
@@ -269,6 +275,6 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
         plt.tight_layout()
 
         print("Generating plot: " + m)
-        plt.savefig(os.path.join(plot_path, m + '.png'))
+        plt.savefig(os.path.join(plot_path, m + '.png'), dpi=300)
         plt.close()
 print("Process finished")

pipelines/hic_tendency.py

@@ -200,7 +200,7 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
         y = np.mean(NormData[m][order[-1]])
         if target_category == 'genes':
             y2 = 0.05 + 0.8 * n / len(these_marks)
-            plt.ylim(0, 0.8)
+            plt.ylim(0, 0.9)
         else:
             y2 = 0.05 + 7 * n / len(these_marks)
             #plt.ylim(0, 6)
@@ -218,6 +218,8 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
         plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=12)
     else:
         plt.ylabel('Normalized histone marks per length of RIs',fontsize=12)
+    plt.xlabel('Division of compartments in the nucleus',fontsize=12, labelpad=10)
+    axe.xaxis.set_label_coords(0.32,-0.2)
     #fig.suptitle('test title', fontsize=20)
     axe.spines['right'].set_visible(False)
     axe.spines['top'].set_visible(False)
@@ -262,6 +264,7 @@ for num, what in enumerate(['Acethylation', 'Methylation'], 1):
             plt.ylabel('Normalized histone marks per CAGE-seq read',fontsize=12)
         else:
             plt.ylabel('Normalized histone marks per length of RIs',fontsize=12)
+        plt.xlabel('Division of compartments in the nucleus',fontsize=12, labelpad=10)
         axe.spines['right'].set_visible(False)
         axe.spines['top'].set_visible(False)
         axe.spines['bottom'].set_visible(False)

pipelines/snakemake/hic_analysis.smk

@@ -13,11 +13,6 @@ marks_list = config["marks_list"]
 LAD_bed = config["LAD_bed"]
 
 
-# If it is the first analysis for a dataset
-sel_marks = os.path.join(selpath, marks_dataset)
-if not os.path.exists(sel_marks):
-    os.makedirs(sel_marks)
-
 # Paths
 listpath = os.path.join(marklistspath, marks_list)
 sourcemarkspath = os.path.join(allpath, cell_type, marks_dataset)