Mercurial > repos > proteore > filter_keywords_values

comparison filter_kw_val.py @ 0:6a45ccfc0e4c 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 18:36:43 -0500
parents
children d29e469b6b20
comparison
equal deleted inserted replaced
-1:000000000000 0:6a45ccfc0e4c
1 import argparse
2 import re
3
4
5 def options():
6 parser = argparse.ArgumentParser()
7 parser.add_argument("-i", "--input", help="Input file", required=True)
8 parser.add_argument("-m", "--match", help="Exact macth")
9 parser.add_argument("--kw", nargs="+", action="append", help="") #
10 parser.add_argument("--kw_file", nargs="+", action="append", help="")
11 parser.add_argument("--value", nargs="+", action="append", help="")
12 parser.add_argument("-o", "--output", default="output.txt")
13 parser.add_argument("--trash_file", default="trash_MQfilter.txt")
14
15 args = parser.parse_args()
16
17 filters(args)
18
19 # python filter2.py -i "/projet/galaxydev/galaxy/tools/proteore_uc1/proteinGroups_Maud.txt" --protein_IDs "A2A288:A8K2U0" --peptides 2 "=" -o "test-data/output_MQfilter.txt"
20
21
22 def isnumber(format, n):
23 float_format = re.compile("^[\-]?[1-9][0-9]*\.?[0-9]+$")
24 int_format = re.compile("^[\-]?[1-9][0-9]*$")
25 test = ""
26 if format == "int":
27 test = re.match(int_format, n)
28 elif format == "float":
29 test = re.match(float_format, n)
30 if test:
31 return True
32 else:
33 return False
34
35 def filters(args):
36 MQfilename = args.input.split(",")[0]
37 header = args.input.split(",")[1]
38 MQfile = readMQ(MQfilename)
39 results = [MQfile, None]
40
41 if args.kw:
42 keywords = args.kw
43 for k in keywords:
44 results = filter_keyword(results[0], header, results[1], k[0], k[1], k[2])
45 if args.kw_file:
46 key_files = args.kw_file
47 for kf in key_files:
48 ids = readOption(kf[0])
49 results = filter_keyword(results[0], header, results[1], ids, kf[1], kf[2])
50 if args.value:
51 for v in args.value:
52 if isnumber("float", v[0]):
53 results = filter_value(results[0], header, results[1], v[0], v[1], v[2])
54 else:
55 raise ValueError("Please enter a number in filter by value")
56
57 # Write results to output
58 output = open(args.output, "w")
59 output.write("".join(results[0]))
60 output.close()
61
62 # Write deleted lines to trash_file
63 trash = open(args.trash_file, "w")
64 #print("".join(results[1]))
65 trash.write("".join(results[1]))
66 trash.close()
67
68 def readOption(filename):
69 f = open(filename, "r")
70 file = f.read()
71 #print(file)
72 filter_list = file.split("\n")
73 #print(filter_list)
74 filters = ""
75 for i in filter_list:
76 filters += i + ":"
77 filters = filters[:-1]
78 #print(filters)
79 return filters
80
81 def readMQ(MQfilename):
82 # Read MQ file
83 mqfile = open(MQfilename, "r")
84 mq = mqfile.readlines()
85 # Remove empty lines (contain only space or new line or "")
86 [mq.remove(blank) for blank in mq if blank.isspace() or blank == ""]
87 return mq
88
89 def filter_keyword(MQfile, header, filtered_lines, ids, ncol, match):
90 mq = MQfile
91 if isnumber("int", ncol.replace("c", "")):
92 id_index = int(ncol.replace("c", "")) - 1 #columns.index("Majority protein IDs")
93 else:
94 raise ValueError("Please specify the column where you would like to apply the filter with valid format")
95
96 ids = ids.upper().split(":")
97 [ids.remove(blank) for blank in ids if blank.isspace() or blank == ""]
98
99 if header == "true":
100 header = mq[0]
101 content = mq[1:]
102 else:
103 header = ""
104 content = mq[:]
105
106 if not filtered_lines: # In case there is already some filtered lines from other filters
107 filtered_lines = []
108 if header != "":
109 filtered_lines.append(header)
110
111 for line in content:
112 id_inline = line.split("\t")[id_index].replace('"', "").split(";")
113 one_id_line = line.replace(line.split("\t")[id_index], id_inline[0]) # Take only first IDs
114
115 if match != "false":
116 # Filter protein IDs
117 if any (pid.upper() in ids for pid in id_inline):
118 #ids = prot_ids.split(":")
119 #print(prot_ids.split(":"))
120 #if prot_id in ids:
121 filtered_lines.append(one_id_line)
122 mq.remove(line)
123 else:
124 mq[mq.index(line)] = one_id_line
125 else:
126 if any (ft in pid.upper() for pid in id_inline for ft in ids):
127 filtered_lines.append(one_id_line)
128 mq.remove(line)
129 else:
130 mq[mq.index(line)] = one_id_line
131 return mq, filtered_lines
132
133 def filter_value(MQfile, header, filtered_prots, filter_value, ncol, opt):
134 mq = MQfile
135 if ncol and isnumber("int", ncol.replace("c", "")): #"Gene names" in columns:
136 index = int(ncol.replace("c", "")) - 1 #columns.index("Gene names")
137 else:
138 raise ValueError("Please specify the column where you would like to apply the filter with valid format")
139
140 if header == "true":
141 header = mq[0]
142 content = mq[1:]
143 else:
144 header = ""
145 content = mq[:]
146
147 if not filtered_prots: # In case there is already some filtered lines from other filters
148 filtered_prots = []
149 if header != "":
150 filtered_prots.append(header)
151
152 for prot in content:
153 filter_value = float(filter_value)
154 pep = prot.split("\t")[index].replace('"', "")
155 if pep.replace(".", "", 1).isdigit():
156 if opt == "<":
157 if not float(pep) < filter_value:
158 filtered_prots.append(prot)
159 mq.remove(prot)
160 elif opt == "<=":
161 if not float(pep) <= filter_value:
162 filtered_prots.append(prot)
163 mq.remove(prot)
164 elif opt == ">":
165 #print(prot.number_of_prots, filter_value, int(prot.number_of_prots) > filter_value)
166 if not float(pep) > filter_value:
167 filtered_prots.append(prot)
168 mq.remove(prot)
169 elif opt == ">=":
170 if not float(pep) >= filter_value:
171 filtered_prots.append(prot)
172 mq.remove(prot)
173 else:
174 if not float(pep) == filter_value:
175 filtered_prots.append(prot)
176 mq.remove(prot)
177 return mq, filtered_prots #output, trash_file
178
179 if __name__ == "__main__":
180 options()