Mercurial > repos > proteore > proteore_protein_interaction_maps

comparison build_protein_interaction_maps.py @ 0:3fcdc585cd2e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 696e05eadff826a980b1150be5db5d1f52d06b1b-dirty
author proteore
date Fri, 01 Mar 2019 11:23:22 -0500
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:3fcdc585cd2e
1 # -*- coding: utf-8 -*-
2 import csv, json, argparse, re
3
4 def get_args() :
5 parser = argparse.ArgumentParser()
6 parser.add_argument("--species")
7 parser.add_argument("--database", help="Humap, Bioplex or Biogrid", required=True)
8 parser.add_argument("--dict_path", required=True)
9 parser.add_argument("--input_type", help="type of input (list of id or filename)",required=True)
10 parser.add_argument("--input", required=True)
11 parser.add_argument("--header")
12 parser.add_argument("--ncol")
13 parser.add_argument("--id_type")
14 parser.add_argument("--network_output")
15 parser.add_argument("--nodes_output")
16 args = parser.parse_args()
17
18 if args.input_type=="file" :
19 args.ncol = nb_col_to_int(args.ncol)
20 args.header = str2bool(args.header)
21
22 return args
23
24 #Turn string into boolean
25 def str2bool(v):
26 if v.lower() in ('yes', 'true', 't', 'y', '1'):
27 return True
28 elif v.lower() in ('no', 'false', 'f', 'n', '0'):
29 return False
30 else:
31 raise argparse.ArgumentTypeError('Boolean value expected.')
32
33 #return the column number in int format
34 def nb_col_to_int(nb_col):
35 try :
36 nb_col = int(nb_col.replace("c", "")) - 1
37 return nb_col
38 except :
39 sys.exit("Please specify the column where you would like to apply the filter with valid format")
40
41 #return list of (unique) ids from string
42 def get_input_ids_from_string(input) :
43 ids_list = list(set(re.split(r'\s+',input.replace("\r","").replace("\n"," ").replace("\t"," "))))
44 if "" in ids_list : ids_list.remove("")
45 #if "NA" in ids_list : ids_list.remove("NA")
46 return ids_list
47
48 #return input_file and list of unique ids from input file path
49 def get_input_ids_from_file(input,nb_col,header) :
50 with open(input, "r") as csv_file :
51 input_file= list(csv.reader(csv_file, delimiter='\t'))
52
53 input_file, ids_list = one_id_one_line(input_file,nb_col,header)
54 if "" in ids_list : ids_list.remove("")
55 #if "NA" in ids_list : ids_list.remove("NA")
56
57 return input_file, ids_list
58
59 #return input file by adding lines when there are more than one id per line
60 def one_id_one_line(input_file,nb_col,header) :
61
62 if header :
63 new_file = [input_file[0]]
64 input_file = input_file[1:]
65 else :
66 new_file=[]
67 ids_list=[]
68
69 for line in input_file :
70 if line != [] and set(line) != {''}:
71 line[nb_col] = re.sub(r"\s+","",line[nb_col])
72 if ";" in line[nb_col] :
73 ids = line[nb_col].split(";")
74 for id in ids :
75 new_file.append(line[:nb_col]+[id]+line[nb_col+1:])
76 ids_list.append(id)
77 else :
78 new_file.append(line)
79 ids_list.append(line[nb_col])
80
81 ids_list= list(set(ids_list))
82
83 return new_file, ids_list
84
85 #replace all blank cells to NA
86 def blank_to_NA(csv_file) :
87 tmp=[]
88 for line in csv_file :
89 line = ["NA" if cell=="" or cell==" " or cell=="NaN" or cell=="-" else cell for cell in line]
90 tmp.append(line)
91
92 return tmp
93
94 def biogrid_output_files(ids,species) :
95 network_file=[["Entrez Gene Interactor A","Entrez Gene Interactor B","Gene symbol Interactor A","Gene symbol Interactor B","Experimental System","Experimental Type","Pubmed ID","Interaction Score","Phenotypes"]]
96 ids_set= set(ids)
97 ids_not_found=set([])
98 for id in ids :
99 if id in ppi_dict['network'] :
100 network_file.extend(ppi_dict['network'][id])
101 ids_set.update([interact[1] for interact in ppi_dict['network'][id]])
102 else :
103 ids_not_found.add(id)
104
105 nodes_file = [["Entrez gene ID","Official Symbol Interactor","Present in user input ids","ID present in Biogrid "+species,"Pathway"]]
106 for id in ids_set:
107 #get pathway
108 if id in ppi_dict['nodes']:
109 description_pathway=";".join(ppi_dict['nodes'][id])
110 else :
111 description_pathway="NA"
112
113 #get gene name
114 if id in ppi_dict['network']: gene_name = ppi_dict['network'][id][0][2]
115 else : gene_name="NA"
116
117 #make line
118 nodes_file.append([id]+[gene_name]+[id in ids]+[id not in ids_not_found]+[description_pathway])
119
120 return network_file,nodes_file
121
122 def bioplex_output_files(ids,id_type,species) :
123 network_file=[[id_type+" Interactor A",id_type+" Interactor B","Gene symbol Interactor A","Gene symbol Interactor B","Interaction Score"]]
124 ids_set= set(ids)
125 ids_not_found=set([])
126 for id in ids :
127 if id in ppi_dict['network'][id_type] :
128 network_file.extend(ppi_dict['network'][id_type][id])
129 ids_set.update([interact[1] for interact in ppi_dict['network'][id_type][id]])
130 else :
131 ids_not_found.add(id)
132
133 if id_type=="UniProt-AC" : nodes_file=[[id_type,"Present in user input ids","ID present in Human Bioplex","Pathway"]]
134 else: nodes_file=[[id_type,"Official symbol Interactor","Present in user input ids","Present in interactome","Pathway"]]
135 for id in ids_set:
136
137 if id in ppi_dict['nodes'][id_type]:
138 description_pathway=";".join(ppi_dict['nodes'][id_type][id])
139 else :
140 description_pathway="NA"
141
142 #make line
143 if id_type=="UniProt-AC":
144 nodes_file.append([id]+[id in ids]+[id not in ids_not_found]+[description_pathway])
145 elif id_type=="GeneID":
146 #get gene_name
147 if id in ppi_dict['network'][id_type]: gene_name = ppi_dict['network'][id_type][id][0][2]
148 else : gene_name="NA"
149 nodes_file.append([id]+[gene_name]+[id in ids]+[id not in ids_not_found]+[description_pathway])
150
151 return network_file,nodes_file
152
153 def humap_output_files(ids,species) :
154 network_file=[["Entrez Gene Interactor A","Entrez Gene Interactor B","Gene symbol Interactor A","Gene symbol Interactor B","Interaction Score"]]
155 ids_set= set(ids)
156 ids_not_found=set([])
157 for id in ids :
158 if id in ppi_dict['network'] :
159 network_file.extend(ppi_dict['network'][id])
160 ids_set.update([interact[1] for interact in ppi_dict['network'][id]])
161 else :
162 ids_not_found.add(id)
163
164 nodes_file = [["Entrez gene ID","Official Symbol Interactor","Present in user input ids","ID present in Hu.MAP","Pathway"]]
165 for id in ids_set:
166 if id in ppi_dict['nodes']:
167 description_pathway=";".join(ppi_dict['nodes'][id])
168 else :
169 description_pathway="NA"
170
171 #get gene name
172 if id in ppi_dict['gene_name']:
173 gene_name = ppi_dict['gene_name'][id]
174 else :
175 gene_name = "NA"
176
177 #make line
178 nodes_file.append([id]+[gene_name]+[id in ids]+[id not in ids_not_found]+[description_pathway])
179
180 return network_file,nodes_file
181
182 #function to sort the csv_file by value in a specific column
183 def sort_by_column(tab,sort_col,reverse,header):
184
185 if len(tab) > 1 : #if there's more than just a header or 1 row
186 if header :
187 head=tab[0]
188 tab=tab[1:]
189
190 #list of empty cells in the column to sort
191 unsortable_lines = [i for i,line in enumerate(tab) if (line[sort_col]=='' or line[sort_col] == 'NA')]
192 unsorted_tab=[ tab[i] for i in unsortable_lines]
193 tab= [line for i,line in enumerate(tab) if i not in unsortable_lines]
194
195 if only_number(tab,sort_col) and any_float(tab,sort_col) :
196 tab = sorted(tab, key=lambda row: float(row[sort_col]), reverse=reverse)
197 elif only_number(tab,sort_col):
198 tab = sorted(tab, key=lambda row: int(row[sort_col]), reverse=reverse)
199 else :
200 tab = sorted(tab, key=lambda row: row[sort_col], reverse=reverse)
201
202 tab.extend(unsorted_tab)
203 if header is True : tab = [head]+tab
204
205 return tab
206
207 def only_number(tab,col) :
208
209 for line in tab :
210 if not (is_number("float",line[col].replace(",",".")) or is_number("int",line[col].replace(",","."))) :
211 return False
212 return True
213
214 #Check if a variable is a float or an integer
215 def is_number(number_format, n):
216 float_format = re.compile(r"^[-]?[0-9][0-9]*.?[0-9]+$")
217 int_format = re.compile(r"^[-]?[0-9][0-9]*$")
218 test = ""
219 if number_format == "int":
220 test = re.match(int_format, n)
221 elif number_format == "float":
222 test = re.match(float_format, n)
223 if test:
224 return True
225
226 #return True is there is at least one float in the column
227 def any_float(tab,col) :
228
229 for line in tab :
230 if is_number("float",line[col].replace(",",".")) :
231 return True
232
233 return False
234
235 def main() :
236
237 #Get args from command line
238 global args
239 args = get_args()
240
241 #get PPI dictionary
242 with open(args.dict_path, 'r') as handle:
243 global ppi_dict
244 ppi_dict = json.load(handle)
245
246 #Get file and/or ids from input
247 if args.input_type == "text" :
248 ids = get_input_ids_from_string(args.input)
249 elif args.input_type == "file" :
250 input_file, ids = get_input_ids_from_file(args.input,args.ncol,args.header)
251
252 #create output files
253 if args.database=="biogrid":
254 network_file, nodes_file = biogrid_output_files(ids,args.species)
255 elif args.database=="bioplex":
256 network_file, nodes_file = bioplex_output_files(ids,args.id_type,args.species)
257 elif args.database=="humap":
258 network_file, nodes_file = humap_output_files(ids,args.species)
259
260 #convert blank to NA and sort files
261 network_file = blank_to_NA(network_file)
262 network_file = sort_by_column(network_file,0,False,True)
263 nodes_file = sort_by_column(nodes_file,0,False,True)
264
265 #write output files
266 with open(args.network_output,"w") as output :
267 writer = csv.writer(output,delimiter="\t")
268 writer.writerows(network_file)
269
270 with open(args.nodes_output,"w") as output :
271 writer = csv.writer(output,delimiter="\t")
272 for row in nodes_file:
273 writer.writerow([unicode(s).encode("utf-8") for s in row])
274
275 if __name__ == "__main__":
276 main()