|
7
|
1 import pandas as pd
|
|
|
2 import matplotlib.patches as mpatches
|
|
|
3 import matplotlib.font_manager as font_manager
|
|
|
4 import matplotlib.pyplot as plt
|
|
13
|
5 import sys
|
|
7
|
6
|
|
|
7 #########################################################################################
|
|
|
8
|
|
|
9 # Read a file and return it as a list
|
|
|
10 def read(path, flag):
|
|
|
11 if flag == 0:
|
|
|
12 with open(path) as fp:
|
|
|
13 file=fp.readlines()
|
|
|
14 fp.close()
|
|
|
15 return file
|
|
|
16
|
|
|
17 if flag == 1:
|
|
|
18 with open(path) as fp:
|
|
|
19 file = fp.read().splitlines()
|
|
|
20 fp.close()
|
|
|
21 return file
|
|
|
22
|
|
|
23 # Write a list to a txt file
|
|
|
24 def write(path, list):
|
|
|
25 with open(path,'w') as fp:
|
|
|
26 for x in list:
|
|
|
27 fp.write(str("\t".join(x[1:-1])))
|
|
|
28 fp.close()
|
|
|
29
|
|
|
30
|
|
|
31 ################################################################################################################################################################>
|
|
|
32
|
|
|
33 def top_diff(miRNA_info, number,flag,l):
|
|
|
34
|
|
|
35 Kind=[]
|
|
|
36
|
|
|
37 miRNA_info.sort(key = lambda x: abs(x[1]),reverse=True)
|
|
|
38 miRNA_info = miRNA_info[:number]
|
|
|
39 miRNA_info.sort(key = lambda x: x[0])
|
|
|
40
|
|
|
41 for x in miRNA_info:
|
|
|
42 if x[1] > 0:
|
|
|
43 Kind.append(True)
|
|
|
44 elif x[1] < 0:
|
|
|
45 Kind.append(False)
|
|
|
46 else:
|
|
|
47 Kind.append("Zero")
|
|
|
48
|
|
|
49 top_miRNA = {"Names": [x[0] for x in miRNA_info],
|
|
|
50 "Log2FC": [x[1] for x in miRNA_info],
|
|
|
51 "Kind": Kind};
|
|
|
52
|
|
|
53 df_miRNA = pd.DataFrame(data=top_miRNA)
|
|
|
54 df_miRNA = df_miRNA.sort_values(by=['Names'])
|
|
|
55 if df_miRNA.empty==False:
|
|
|
56 h1=df_miRNA.plot.barh(x= 'Names',y='Log2FC',color=df_miRNA.Kind.map({True: 'g', False: 'r', 'Zero':'k'}))
|
|
|
57 figure = plt.gcf() # get current figure
|
|
|
58 figure.set_size_inches(5, 12) # set figure's size manually to your full screen (32x18)
|
|
|
59 up_reg = mpatches.Patch(color='green', label='Upregulated')
|
|
|
60 down_reg = mpatches.Patch(color='red', label='Downregulated')
|
|
|
61 font = font_manager.FontProperties(weight='bold', style='normal')
|
|
|
62 l3 = plt.legend(handles=[up_reg,down_reg],bbox_to_anchor=(1.04,0.5), loc="center left", borderaxespad=0)
|
|
|
63 h1.set_ylabel(" ", fontsize=3, fontweight='bold')
|
|
|
64 h1.set_xlabel("Log2FC", fontsize=12, fontweight='bold')
|
|
|
65 plt.axvline(x=0, color="k")
|
|
|
66
|
|
|
67 plt.grid(axis='y', linewidth=0.2)
|
|
|
68 plt.grid(axis='x', linewidth=0.2)
|
|
|
69 if flag=='t':
|
|
|
70 plt.savefig('tem.png', bbox_inches='tight', dpi=300)
|
|
|
71 if flag=='nt':
|
|
|
72 plt.savefig('non.png', bbox_inches='tight', dpi=300)
|
|
|
73
|
|
|
74
|
|
|
75 ################################################################################################################################################################>
|
|
|
76
|
|
|
77 def unique(sequence):
|
|
|
78 seen = set()
|
|
|
79 return [x for x in sequence if not (x in seen or seen.add(x))]
|
|
|
80
|
|
|
81 ################################################################################################################################################################>
|
|
|
82
|
|
|
83 def top_scatter_non(matures,isoforms,non_temp,uni_names,number):
|
|
|
84
|
|
|
85 mat_names=[]
|
|
|
86 mat_log2fc=[]
|
|
|
87
|
|
|
88 iso_names=[]
|
|
|
89 iso_log2fc=[]
|
|
|
90
|
|
|
91 non_temp_names=[]
|
|
|
92 non_temp_log2fc=[]
|
|
|
93
|
|
|
94 count=0
|
|
|
95 for x in uni_names:
|
|
|
96 flag = False
|
|
|
97 if count<number:
|
|
|
98 for y in matures:
|
|
|
99 if x in y[0]:
|
|
|
100 mat_log2fc.append(y[1])
|
|
|
101 mat_names.append(x)
|
|
|
102 flag=True
|
|
|
103 for y in isoforms:
|
|
|
104 if x in y[0]:
|
|
|
105 iso_log2fc.append(y[1])
|
|
|
106 iso_names.append(x)
|
|
|
107 flag=True
|
|
|
108 for y in non_temp:
|
|
|
109 if x in y[0]:
|
|
|
110 non_temp_log2fc.append(y[1])
|
|
|
111 non_temp_names.append(x)
|
|
|
112 flag=True
|
|
|
113 if flag==True:
|
|
|
114 count+=1
|
|
|
115
|
|
|
116 mat_df = pd.DataFrame(dict(names=mat_names, log2fc=mat_log2fc))
|
|
|
117 iso_df = pd.DataFrame(dict(names=iso_names, log2fc=iso_log2fc))
|
|
|
118 non_df = pd.DataFrame(dict(names=non_temp_names, log2fc= non_temp_log2fc))
|
|
|
119
|
|
|
120 iso_df.sort_values(by=['names'])
|
|
|
121 mat_df.sort_values(by=['names'])
|
|
|
122 non_df.sort_values(by=['names'])
|
|
|
123
|
|
|
124 fig, ax = plt.subplots()
|
|
|
125
|
|
|
126 h3=ax.scatter(iso_df['log2fc'],iso_df['names'],edgecolors='k',linewidth=1, marker='o', c='red',alpha=0.4)
|
|
|
127 h1=ax.scatter(mat_df['log2fc'],mat_df['names'],edgecolors='k',linewidth=1, marker='o', c='green',alpha=0.4)
|
|
|
128 h2=ax.scatter(non_df['log2fc'],non_df['names'],edgecolors='k',linewidth=1, marker='o', c='orange',alpha=0.4)
|
|
|
129
|
|
|
130 l3 = plt.legend([h1,h2,h3],["RefSeq miRNA","Non-templated isomiR","Templated isomiR"],bbox_to_anchor=(1.04,0.5), loc="center left", borderaxespad=0)
|
|
|
131 plt.axvline(x=0, color="k")
|
|
|
132 plt.grid(axis='y', linewidth=0.2)
|
|
|
133 plt.grid(axis='x', linewidth=0.2)
|
|
|
134 plt.xlabel("Log2FC", fontsize=12, fontweight='bold')
|
|
|
135 plt.yticks(rotation=0,ha="right", fontsize=10)
|
|
|
136 plt.xticks(rotation=0,ha="right", fontsize=10)
|
|
|
137 plt.tight_layout()
|
|
|
138 figure = plt.gcf() # get current figure
|
|
|
139 figure.set_size_inches(16, 12) # set figure's size manually to your full screen (32x18)
|
|
|
140 plt.savefig('a2.png', bbox_inches='tight', dpi=300)
|
|
|
141
|
|
|
142 #########################################################################################################################################################################################################################################
|
|
|
143
|
|
|
144 def top_scatter_tem(matures,isoforms,uni_names,number):
|
|
|
145
|
|
|
146 mat_names=[]
|
|
|
147 mat_log2fc=[]
|
|
|
148
|
|
|
149 iso_names=[]
|
|
|
150 iso_log2fc=[]
|
|
|
151
|
|
|
152 count=0
|
|
|
153 for x in uni_names:
|
|
|
154 flag = False
|
|
|
155 if count<number:
|
|
|
156 for y in matures:
|
|
|
157 if x in y[0]:
|
|
|
158 mat_log2fc.append(y[1])
|
|
|
159 mat_names.append(x)
|
|
|
160 flag=True
|
|
|
161 for y in isoforms:
|
|
|
162 if x in y[0]:
|
|
|
163 iso_log2fc.append(y[1])
|
|
|
164 iso_names.append(x)
|
|
|
165 flag=True
|
|
|
166 if flag==True:
|
|
|
167 count+=1
|
|
|
168
|
|
|
169 mat_df = pd.DataFrame(dict(names=mat_names, log2fc=mat_log2fc))
|
|
|
170 iso_df = pd.DataFrame(dict(names=iso_names, log2fc=iso_log2fc))
|
|
|
171
|
|
|
172 iso_df.sort_values(by=['names'])
|
|
|
173 mat_df.sort_values(by=['names'])
|
|
|
174
|
|
|
175 fig, ax = plt.subplots()
|
|
|
176
|
|
|
177 h3=ax.scatter(iso_df['log2fc'],iso_df['names'],edgecolors='k',linewidth=1, marker='o', c='red',alpha=0.4)
|
|
|
178 h1=ax.scatter(mat_df['log2fc'],mat_df['names'],edgecolors='k',linewidth=1, marker='o', c='green',alpha=0.4)
|
|
|
179
|
|
|
180 l3 = plt.legend([h1,h3],["RefSeq miRNA","Templated isomiR"],bbox_to_anchor=(1.04,0.5), loc="center left", borderaxespad=0)
|
|
|
181 plt.axvline(x=0, color="k")
|
|
|
182 plt.grid(axis='y', linewidth=0.2)
|
|
|
183 plt.grid(axis='x', linewidth=0.2)
|
|
|
184 plt.xlabel("Log2FC", fontsize=12, fontweight='bold')
|
|
|
185 plt.yticks(rotation=0,ha="right", fontsize=10)
|
|
|
186 plt.xticks(rotation=0,ha="right", fontsize=10)
|
|
|
187 plt.tight_layout()
|
|
|
188 figure = plt.gcf() # get current figure
|
|
|
189 figure.set_size_inches(16, 12) # set figure's size manually to your full screen (32x18)
|
|
|
190 plt.savefig('a2.png', bbox_inches='tight', dpi=300)
|
|
|
191
|
|
|
192
|
|
|
193 ##############################################################################################################################################################################################################################################
|
|
|
194
|
|
|
195 def preproccess(non_templated,matures,isoforms,log2fc,pval,stat):
|
|
|
196
|
|
|
197 if stat=="3":
|
|
|
198 non_temp = [[x[0],float(x[1]),float(x[2])] for x in non_templated if abs(float(x[1]))>log2fc and float(x[2])>pval]
|
|
|
199 mat = [[x[0],float(x[1]),float(x[2])] for x in matures if abs(float(x[1]))>log2fc and float(x[2])>pval]
|
|
|
200 iso = [[x[0],float(x[1]),float(x[2])] for x in isoforms if abs(float(x[1]))>log2fc and float(x[2])>pval]
|
|
|
201 else:
|
|
|
202 non_temp = [[x[0],float(x[1]),float(x[2])] for x in non_templated if abs(float(x[1]))>log2fc and float(x[2])<pval]
|
|
|
203 mat = [[x[0],float(x[1]),float(x[2])] for x in matures if abs(float(x[1]))>log2fc and float(x[2])<pval]
|
|
|
204 iso = [[x[0],float(x[1]),float(x[2])] for x in isoforms if abs(float(x[1]))>log2fc and float(x[2])<pval]
|
|
|
205
|
|
|
206 mat_iso = mat+iso
|
|
|
207
|
|
|
208 if not non_temp and not mat and not iso:
|
|
|
209 sys.exit("There aren't entries which meet these criteria")
|
|
|
210
|
|
|
211 mat.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
212 iso.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
213 non_temp.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
214
|
|
|
215 all=mat+iso+non_temp
|
|
|
216 all.sort(key = lambda x: abs(float(x[1])), reverse=True)
|
|
|
217 names=[x[0].split("_")[0] for x in all]
|
|
|
218 uni_names=unique(names)
|
|
|
219
|
|
|
220 diff_non_templated = [[x[0],float(x[1]),float(x[2])] for x in non_templated if abs(float(x[1]))>1 and float(x[2])<pval and x[0].split("_")[0] in uni_names]
|
|
|
221 diff_matures = [[x[0],float(x[1]),float(x[2])] for x in matures if abs(float(x[1]))>1 and float(x[2])<pval and x[0].split("_")[0] in uni_names]
|
|
|
222 diff_isoforms = [[x[0],float(x[1]),float(x[2])] for x in isoforms if abs(float(x[1]))>1 and float(x[2])<pval and x[0].split("_")[0] in uni_names]
|
|
|
223
|
|
|
224 diff_matures.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
225 diff_isoforms.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
226 diff_non_templated.sort(key = lambda x: abs(float(x[1])),reverse=True)
|
|
|
227
|
|
|
228 return diff_matures,diff_isoforms,diff_non_templated,uni_names,non_temp,mat_iso
|
|
|
229
|
|
|
230 ################################################################################################################################################################################################################################################>
|
|
|
231
|
