Mercurial > repos > proteore > proteore_venn_diagram

comparison venn_diagram.py @ 0:9d4f20618ab4 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit abb24d36c776520e73220d11386252d848173697-dirty
author proteore
date Sun, 26 Nov 2017 19:09:40 -0500
parents
children 145f347dc0e1
comparison
equal deleted inserted replaced
-1:000000000000 0:9d4f20618ab4
1 #!/usr/bin/env python2.7
2
3 import os
4 import sys
5 import json
6 import operator
7 import argparse
8 import re
9 from itertools import combinations
10
11 CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
12
13 ##################################################################################################################################################
14 # FUNCTIONS
15 ##################################################################################################################################################
16
17 def isnumber(format, n):
18 """
19 Check if an element is integer or float
20 """
21 float_format = re.compile("^[\-]?[1-9][0-9]*\.?[0-9]+$")
22 int_format = re.compile("^[\-]?[1-9][0-9]*$")
23 test = ""
24 if format == "int":
25 test = re.match(int_format, n)
26 elif format == "float":
27 test = re.match(float_format, n)
28 if test:
29 return True
30 else:
31 return False
32
33 def input_to_dict(inputs):
34 """
35 Parse input and return a dictionary of name and data of each lists/files
36 """
37 comp_dict = {}
38 title_dict = {}
39 c = ["A", "B", "C", "D", "E", "F"]
40 for i in range(len(inputs)):
41 input_file = inputs[i][0]
42 name = inputs[i][1]
43 input_type = inputs[i][2]
44 title = c[i]
45 title_dict[title] = name
46 ids = set()
47 if input_type == "file":
48 header = inputs[i][3]
49 ncol = inputs[i][4]
50 file_content = open(input_file, "r").readlines()
51
52 # Check if column number is in right form
53 if isnumber("int", ncol.replace("c", "")):
54 if header == "true":
55 file_content = [x.strip() for x in [line.split("\t")[int(ncol.replace("c", ""))-1].split(";")[0] for line in file_content[1:]]] # take only first IDs
56 else:
57 file_content = [x.strip() for x in [line.split("\t")[int(ncol.replace("c", ""))-1].split(";")[0] for line in file_content]] # take only first IDs
58 #print(file_content[1:13])
59 else:
60 raise ValueError("Please fill in the right format of column number")
61 else:
62 ids = set()
63 file_content = inputs[i][0].split()
64
65 ids.update(file_content)
66 comp_dict[title] = ids
67
68 return comp_dict, title_dict
69
70 def intersect(comp_dict):
71 """
72 Calculate the intersections of input
73 """
74 names = set(comp_dict)
75 for i in range(1, len(comp_dict) + 1):
76 for group in combinations(sorted(comp_dict), i):
77 others = set()
78 [others.add(name) for name in names if name not in group]
79 difference = []
80 intersected = set.intersection(*(comp_dict[k] for k in group))
81 n = "".join(group)
82 if len(others) > 0:
83 difference = intersected.difference(set.union(*(comp_dict[k] for k in others)))
84 yield group, list(intersected), list(difference)
85
86 def diagram(comp_dict, title_dict):
87 """
88 Create json string for jvenn diagram plot
89 """
90 result = {}
91 result["name"] = {}
92 for k in comp_dict.keys():
93 result["name"][k] = title_dict[k]
94
95 result["data"] = {}
96 result["values"] = {}
97 for group, intersected, difference in intersect(comp_dict):
98 if len(group) == 1:
99 result["data"]["".join(group)] = difference
100 result["values"]["".join(group)] = len(difference)
101 elif len(group) > 1 and len(group) < len(comp_dict):
102 result["data"]["".join(group)] = difference
103 result["values"]["".join(group)] = len(difference)
104 elif len(group) == len(comp_dict):
105 result["data"]["".join(group)] = intersected
106 result["values"]["".join(group)] = len(intersected)
107
108 return result
109
110 def write_text_venn(json_result):
111 """
112 Write intersections of input to text output file
113 """
114 output = open("venn_diagram_text_output.txt", "w")
115 string = ""
116 lines = []
117 result = dict((k, v) for k, v in json_result["data"].iteritems() if v != [])
118 print(result)
119 max_count = max(len(v) for v in result.values())
120 print(max_count)
121 for i in range(max_count):
122 lines.append("")
123
124 for i in range(max_count):
125 header = ""
126 for d in range(len(result.keys())):
127 data = result.keys()[d]
128 name = "_".join([json_result["name"][x] for x in data])
129 header += name + "\t"
130 if len(result[data]) > i:
131 lines[i] += result[data][i] + "\t"
132 else:
133 lines[i] += "\t"
134 string += header + "\n"
135 string += "\n".join(lines)
136 print(string)
137 output.write(string)
138 output.close()
139
140 def write_summary( summary_file, inputs):
141 """
142 Paste json string into template file
143 """
144 a, b = input_to_dict(inputs)
145 data = diagram(a, b)
146 write_text_venn(data)
147
148 to_replace = {
149 "series": [data],
150 "displayStat": "true",
151 "displaySwitch": "true",
152 "shortNumber": "true",
153 }
154
155 FH_summary_tpl = open(os.path.join(CURRENT_DIR, "jvenn_template.html"))
156 FH_summary_out = open(summary_file, "w" )
157 for line in FH_summary_tpl:
158 if "###JVENN_DATA###" in line:
159 line = line.replace("###JVENN_DATA###", json.dumps(to_replace))
160 FH_summary_out.write(line)
161
162 FH_summary_out.close()
163 FH_summary_tpl.close()
164
165 def process(args):
166 write_summary(args.summary, args.input)
167
168
169 ##################################################################################################################################################
170 # MAIN
171 ##################################################################################################################################################
172 if __name__ == '__main__':
173 # Parse parameters
174 parser = argparse.ArgumentParser(description='Filters an abundance file')
175 group_input = parser.add_argument_group( 'Inputs' )
176 group_input.add_argument('--input', nargs="+", action="append", required=True, help="The input tabular file.")
177 group_output = parser.add_argument_group( 'Outputs' )
178 group_output.add_argument('--summary', default="summary.html", help="The HTML file containing the graphs. [Default: %(default)s]")
179 args = parser.parse_args()
180
181 # Process
182 process( args )