generate boxplot for methylation frenquency

boxplot_methylation_frequency.py

+import pandas as pd
+import matplotlib.pyplot as plt
+import numpy as np
+
+# Reading files:
+bed1 = pd.read_csv("C:\\Users\\SCD UM\\Desktop\\methylation_frequency\\MET1_DSP.bed", sep='\t', names=['chro', 'start', 'end', 'freq'])
+bed2 = pd.read_csv("C:\\Users\\SCD UM\\Desktop\\methylation_frequency\\MET1_Tombo.bed", sep='\t', names=['chro', 'start', 'end', 'freq'])
+bed3 = pd.read_csv("C:\\Users\\SCD UM\\Desktop\\methylation_frequency\\MET1_Megalodon.bed", sep='\t', names=['chro', 'start', 'end', 'freq'])
+bed4 = pd.read_csv("C:\\Users\\SCD UM\\Desktop\\methylation_frequency\\MET1_DeepMP.bed", sep='\t', names=['chro', 'start', 'end','freq'])
+
+freq1 = bed1["freq"]
+freq2 = bed2["freq"]
+freq3 = bed3["freq"]
+freq4 = bed4["freq"]
+
+
+# Sorting frequency data
+freq1_sorted = np.sort(freq1)
+freq2_sorted = np.sort(freq2)
+freq3_sorted = np.sort(freq3)
+freq4_sorted = np.sort(freq4)
+
+# Dividing data based on frequency ranges
+
+# Bed 1
+freq1_low = freq1_sorted[(freq1_sorted >= 0.25) & (freq1_sorted < 0.5)]
+freq1_medium = freq1_sorted[(freq1_sorted >= 0.5) & (freq1_sorted < 0.75)]
+freq1_high = freq1_sorted[freq1_sorted >= 0.75]
+
+# Bed 2
+freq2_low = freq2_sorted[(freq2_sorted >= 0.25) & (freq2_sorted < 0.5)]
+freq2_medium = freq2_sorted[(freq2_sorted >= 0.5) & (freq2_sorted < 0.75)]
+freq2_high = freq2_sorted[freq2_sorted >= 0.75]
+
+# Bed 3
+freq3_low = freq3_sorted[(freq3_sorted >= 0.25) & (freq3_sorted < 0.5)]
+freq3_medium = freq3_sorted[(freq3_sorted >= 0.5) & (freq3_sorted < 0.75)]
+freq3_high = freq3_sorted[freq3_sorted >= 0.75]
+
+# Bed 4
+freq4_low = freq4_sorted[(freq4_sorted >= 0.25) & (freq4_sorted < 0.5)]
+freq4_medium = freq4_sorted[(freq4_sorted >= 0.5) & (freq4_sorted < 0.75)]
+freq4_high = freq4_sorted[freq4_sorted >= 0.75]
+
+
+# Creating subplots
+fig, axes = plt.subplots(nrows=1, ncols=3, figsize=(12, 4))
+
+# Color definitions
+color_bed1 = 'lightblue'
+color_bed2 = 'lightgreen'
+color_bed3 = 'red'
+color_bed4 = 'pink'
+
+# Plotting boxplots for each frequency range
+boxes_low = axes[0].boxplot([freq1_low, freq2_low, freq3_low, freq4_low], patch_artist=True, labels=["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"])
+axes[0].set_xticklabels(["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"], rotation=45)
+boxes_medium = axes[1].boxplot([freq1_medium, freq2_medium, freq3_medium, freq4_medium], patch_artist=True, labels=["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"])
+axes[1].set_xticklabels(["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"], rotation=45)
+boxes_high = axes[2].boxplot([freq1_high, freq2_high, freq3_high, freq4_high], patch_artist=True, labels=["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"])
+axes[2].set_xticklabels(["DeepSignalPlant", "Tombo", "Megalodon", "DeepMP"], rotation=45)
+
+
+# Setting colors for the boxes
+for i, box in enumerate(boxes_low['boxes']):
+    if i == 0:
+        box.set(facecolor=color_bed1)
+    elif i == 1:
+        box.set(facecolor=color_bed2)
+    elif i == 2:
+        box.set(facecolor=color_bed3)
+    elif i == 3:
+        box.set(facecolor=color_bed4)
+
+for i, box in enumerate(boxes_medium['boxes']):
+    if i == 0:
+        box.set(facecolor=color_bed1)
+    elif i == 1:
+        box.set(facecolor=color_bed2)
+    elif i == 2:
+        box.set(facecolor=color_bed3)
+    elif i == 3:
+        box.set(facecolor=color_bed4)
+
+for i, box in enumerate(boxes_high['boxes']):
+    if i == 0:
+        box.set(facecolor=color_bed1)
+    elif i == 1:
+        box.set(facecolor=color_bed2)
+    elif i == 2:
+        box.set(facecolor=color_bed3)
+    elif i == 3:
+        box.set(facecolor=color_bed4)
+
+# Setting titles and y-axis labels
+axes[0].set_title("Faible méthylation (0.25 - 0.5)")
+axes[0].set_ylabel("Fréquence de méthylation")
+axes[1].set_title("Moyenne méthylation (0.5 - 0.75)")
+axes[1].set_ylabel("Fréquence de méthylation")
+axes[2].set_title("Forte méthylation (0.75-1.0)")
+axes[2].set_ylabel("Fréquence de méthylation")
+
+# Setting the same y-axis limits for all subplots
+# Calculating the overall minimum and maximum values
+min_value = min(np.min(freq1_low), np.min(freq2_low), np.min(freq3_low), np.min(freq4_low))
+max_value = max(np.max(freq1_high), np.max(freq2_high), np.max(freq3_high), np.max(freq4_high))
+
+# Setting the same y-axis limits for all subplots with additional space
+buffer = 0.05  # Adjust this value to increase or decrease the buffer space
+y_min = min_value - buffer * (max_value - min_value)
+y_max = max_value + buffer * (max_value - min_value)
+
+for ax in axes:
+    ax.set_ylim(y_min, y_max)
+
+# Adjusting the layout
+plt.tight_layout()
+
+# Display the plot
+plt.show()