|
47
|
1 import argparse
|
|
|
2 import logging
|
|
|
3 import sys
|
|
|
4 import os
|
|
|
5 import re
|
|
|
6
|
|
0
|
7 from collections import defaultdict
|
|
47
|
8
|
|
|
9 def main():
|
|
|
10 parser = argparse.ArgumentParser()
|
|
|
11 parser.add_argument("--input", help="The '7_V-REGION-mutation-and-AA-change-table' and '10_V-REGION-mutation-hotspots' merged together, with an added 'best_match' annotation")
|
|
|
12 parser.add_argument("--genes", help="The genes available in the 'best_match' column")
|
|
|
13 parser.add_argument("--empty_region_filter", help="Where does the sequence start?", choices=['leader', 'FR1', 'CDR1', 'FR2'])
|
|
|
14 parser.add_argument("--output", help="Output file")
|
|
0
|
15
|
|
47
|
16 args = parser.parse_args()
|
|
|
17
|
|
|
18 infile = args.input
|
|
|
19 genes = str(args.genes).split(",")
|
|
|
20 empty_region_filter = args.empty_region_filter
|
|
|
21 outfile = args.output
|
|
0
|
22
|
|
47
|
23 genedic = dict()
|
|
0
|
24
|
|
47
|
25 mutationdic = dict()
|
|
62
|
26 mutationMatcher = re.compile("^(.)(\d+).(.),?[ ]?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?")
|
|
|
27 mutationMatcher = re.compile("^([actg])(\d+).([actg]),?[ ]?([A-Z])?(\d+)?.?([A-Z])?(.*)?")
|
|
|
28 mutationMatcher = re.compile("^([actg])(\d+).([actg]),?[ ]?([A-Z])?(\d+)?[>]?([A-Z;])?(.*)?")
|
|
|
29 mutationMatcher = re.compile("^([nactg])(\d+).([nactg]),?[ ]?([A-Z])?(\d+)?[>]?([A-Z;])?(.*)?")
|
|
47
|
30 NAMatchResult = (None, None, None, None, None, None, '')
|
|
|
31 geneMatchers = {gene: re.compile("^" + gene + ".*") for gene in genes}
|
|
|
32 linecount = 0
|
|
0
|
33
|
|
47
|
34 IDIndex = 0
|
|
|
35 best_matchIndex = 0
|
|
|
36 fr1Index = 0
|
|
|
37 cdr1Index = 0
|
|
|
38 fr2Index = 0
|
|
|
39 cdr2Index = 0
|
|
|
40 fr3Index = 0
|
|
|
41 first = True
|
|
|
42 IDlist = []
|
|
|
43 mutationList = []
|
|
|
44 mutationListByID = {}
|
|
|
45 cdr1LengthDic = {}
|
|
|
46 cdr2LengthDic = {}
|
|
|
47
|
|
|
48 fr1LengthDict = {}
|
|
|
49 fr2LengthDict = {}
|
|
|
50 fr3LengthDict = {}
|
|
|
51
|
|
|
52 cdr1LengthIndex = 0
|
|
|
53 cdr2LengthIndex = 0
|
|
0
|
54
|
|
47
|
55 fr1SeqIndex = 0
|
|
|
56 fr2SeqIndex = 0
|
|
|
57 fr3SeqIndex = 0
|
|
|
58
|
|
|
59 tandem_sum_by_class = defaultdict(int)
|
|
|
60 expected_tandem_sum_by_class = defaultdict(float)
|
|
0
|
61
|
|
47
|
62 with open(infile, 'ru') as i:
|
|
|
63 for line in i:
|
|
|
64 if first:
|
|
|
65 linesplt = line.split("\t")
|
|
|
66 IDIndex = linesplt.index("Sequence.ID")
|
|
|
67 best_matchIndex = linesplt.index("best_match")
|
|
|
68 fr1Index = linesplt.index("FR1.IMGT")
|
|
|
69 cdr1Index = linesplt.index("CDR1.IMGT")
|
|
|
70 fr2Index = linesplt.index("FR2.IMGT")
|
|
|
71 cdr2Index = linesplt.index("CDR2.IMGT")
|
|
|
72 fr3Index = linesplt.index("FR3.IMGT")
|
|
|
73 cdr1LengthIndex = linesplt.index("CDR1.IMGT.seq")
|
|
|
74 cdr2LengthIndex = linesplt.index("CDR2.IMGT.seq")
|
|
|
75 fr1SeqIndex = linesplt.index("FR1.IMGT.seq")
|
|
|
76 fr2SeqIndex = linesplt.index("FR2.IMGT.seq")
|
|
|
77 fr3SeqIndex = linesplt.index("FR3.IMGT.seq")
|
|
|
78 first = False
|
|
|
79 continue
|
|
|
80 linecount += 1
|
|
0
|
81 linesplt = line.split("\t")
|
|
47
|
82 ID = linesplt[IDIndex]
|
|
|
83 genedic[ID] = linesplt[best_matchIndex]
|
|
62
|
84
|
|
|
85 mutationdic[ID + "_FR1"] = []
|
|
|
86 if len(linesplt[fr1Index]) > 5 and empty_region_filter == "leader":
|
|
|
87 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x]
|
|
|
88
|
|
|
89 mutationdic[ID + "_CDR1"] = []
|
|
|
90 if len(linesplt[cdr1Index]) > 5 and empty_region_filter in ["leader", "FR1"]:
|
|
|
91 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x]
|
|
|
92
|
|
|
93 mutationdic[ID + "_FR2"] = []
|
|
|
94 if len(linesplt[fr2Index]) > 5 and empty_region_filter in ["leader", "FR1", "CDR1"]:
|
|
|
95 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x]
|
|
|
96
|
|
|
97 mutationdic[ID + "_CDR2"] = []
|
|
|
98 if len(linesplt[cdr2Index]) > 5:
|
|
|
99 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x]
|
|
|
100
|
|
|
101 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
|
|
|
102
|
|
|
103 mutationdic[ID + "_FR3"] = []
|
|
|
104 if len(linesplt[fr3Index]) > 5:
|
|
|
105 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x]
|
|
|
106
|
|
47
|
107 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
|
|
|
108 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
|
|
|
109
|
|
|
110 cdr1Length = len(linesplt[cdr1LengthIndex])
|
|
|
111 cdr2Length = len(linesplt[cdr2LengthIndex])
|
|
|
112
|
|
|
113 #print linesplt[fr2SeqIndex]
|
|
|
114 fr1Length = len(linesplt[fr1SeqIndex]) if empty_region_filter == "leader" else 0
|
|
|
115 fr2Length = len(linesplt[fr2SeqIndex]) if empty_region_filter in ["leader", "FR1", "CDR1"] else 0
|
|
|
116 fr3Length = len(linesplt[fr3SeqIndex])
|
|
|
117
|
|
|
118 cdr1LengthDic[ID] = cdr1Length
|
|
|
119 cdr2LengthDic[ID] = cdr2Length
|
|
|
120
|
|
|
121 fr1LengthDict[ID] = fr1Length
|
|
|
122 fr2LengthDict[ID] = fr2Length
|
|
|
123 fr3LengthDict[ID] = fr3Length
|
|
|
124
|
|
|
125 IDlist += [ID]
|
|
62
|
126 print "len(mutationdic) =", len(mutationdic)
|
|
47
|
127
|
|
62
|
128 with open(os.path.join(os.path.dirname(os.path.abspath(infile)), "mutationdict.txt"), 'w') as out_handle:
|
|
|
129 for ID, lst in mutationdic.iteritems():
|
|
|
130 for mut in lst:
|
|
|
131 out_handle.write("{0}\t{1}\n".format(ID, "\t".join([str(x) for x in mut])))
|
|
47
|
132
|
|
|
133 #tandem mutation stuff
|
|
|
134 tandem_frequency = defaultdict(int)
|
|
|
135 mutation_frequency = defaultdict(int)
|
|
48
|
136
|
|
|
137 mutations_by_id_dic = {}
|
|
|
138 first = True
|
|
|
139 mutation_by_id_file = os.path.join(os.path.dirname(outfile), "mutation_by_id.txt")
|
|
|
140 with open(mutation_by_id_file, 'r') as mutation_by_id:
|
|
|
141 for l in mutation_by_id:
|
|
|
142 if first:
|
|
|
143 first = False
|
|
|
144 continue
|
|
|
145 splt = l.split("\t")
|
|
|
146 mutations_by_id_dic[splt[0]] = int(splt[1])
|
|
|
147
|
|
47
|
148 tandem_file = os.path.join(os.path.dirname(outfile), "tandems_by_id.txt")
|
|
|
149 with open(tandem_file, 'w') as o:
|
|
|
150 highest_tandem_length = 0
|
|
|
151
|
|
|
152 o.write("Sequence.ID\tnumber_of_mutations\tnumber_of_tandems\tregion_length\texpected_tandems\tlongest_tandem\ttandems\n")
|
|
|
153 for ID in IDlist:
|
|
|
154 mutations = mutationListByID[ID]
|
|
|
155 if len(mutations) == 0:
|
|
|
156 continue
|
|
|
157 last_mut = max(mutations, key=lambda x: int(x[1]))
|
|
|
158
|
|
|
159 last_mut_pos = int(last_mut[1])
|
|
|
160
|
|
|
161 mut_positions = [False] * (last_mut_pos + 1)
|
|
|
162
|
|
|
163 for mutation in mutations:
|
|
|
164 frm, where, to, frmAA, whereAA, toAA, thing = mutation
|
|
|
165 where = int(where)
|
|
|
166 mut_positions[where] = True
|
|
|
167
|
|
|
168 tandem_muts = []
|
|
|
169 tandem_start = -1
|
|
|
170 tandem_length = 0
|
|
|
171 for i in range(len(mut_positions)):
|
|
|
172 if mut_positions[i]:
|
|
|
173 if tandem_start == -1:
|
|
|
174 tandem_start = i
|
|
|
175 tandem_length += 1
|
|
|
176 #print "".join(["1" if x else "0" for x in mut_positions[:i+1]])
|
|
|
177 else:
|
|
|
178 if tandem_length > 1:
|
|
|
179 tandem_muts.append((tandem_start, tandem_length))
|
|
|
180 #print "{0}{1} {2}:{3}".format(" " * (i - tandem_length), "^" * tandem_length, tandem_start, tandem_length)
|
|
|
181 tandem_start = -1
|
|
|
182 tandem_length = 0
|
|
|
183 if tandem_length > 1: # if the sequence ends with a tandem mutation
|
|
|
184 tandem_muts.append((tandem_start, tandem_length))
|
|
|
185
|
|
|
186 if len(tandem_muts) > 0:
|
|
|
187 if highest_tandem_length < len(tandem_muts):
|
|
|
188 highest_tandem_length = len(tandem_muts)
|
|
0
|
189
|
|
47
|
190 region_length = fr1LengthDict[ID] + cdr1LengthDic[ID] + fr2LengthDict[ID] + cdr2LengthDic[ID] + fr3LengthDict[ID]
|
|
|
191 longest_tandem = max(tandem_muts, key=lambda x: x[1]) if len(tandem_muts) else (0, 0)
|
|
48
|
192 num_mutations = mutations_by_id_dic[ID] # len(mutations)
|
|
47
|
193 f_num_mutations = float(num_mutations)
|
|
|
194 num_tandem_muts = len(tandem_muts)
|
|
|
195 expected_tandem_muts = f_num_mutations * (f_num_mutations - 1.0) / float(region_length)
|
|
|
196 o.write("{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\n".format(ID,
|
|
|
197 str(num_mutations),
|
|
|
198 str(num_tandem_muts),
|
|
|
199 str(region_length),
|
|
|
200 str(round(expected_tandem_muts, 2)),
|
|
|
201 str(longest_tandem[1]),
|
|
|
202 str(tandem_muts)))
|
|
|
203 gene = genedic[ID]
|
|
|
204 if gene.find("unmatched") == -1:
|
|
|
205 tandem_sum_by_class[gene] += num_tandem_muts
|
|
|
206 expected_tandem_sum_by_class[gene] += expected_tandem_muts
|
|
0
|
207
|
|
47
|
208 tandem_sum_by_class["all"] += num_tandem_muts
|
|
|
209 expected_tandem_sum_by_class["all"] += expected_tandem_muts
|
|
0
|
210
|
|
47
|
211 gene = gene[:3]
|
|
|
212 if gene in ["IGA", "IGG"]:
|
|
|
213 tandem_sum_by_class[gene] += num_tandem_muts
|
|
|
214 expected_tandem_sum_by_class[gene] += expected_tandem_muts
|
|
|
215 else:
|
|
|
216 tandem_sum_by_class["unmatched"] += num_tandem_muts
|
|
|
217 expected_tandem_sum_by_class["unmatched"] += expected_tandem_muts
|
|
|
218
|
|
40
|
219
|
|
47
|
220 for tandem_mut in tandem_muts:
|
|
|
221 tandem_frequency[str(tandem_mut[1])] += 1
|
|
|
222 #print "\t".join([ID, str(len(tandem_muts)), str(longest_tandem[1]) , str(tandem_muts)])
|
|
|
223
|
|
|
224 tandem_freq_file = os.path.join(os.path.dirname(outfile), "tandem_frequency.txt")
|
|
|
225 with open(tandem_freq_file, 'w') as o:
|
|
|
226 for frq in sorted([int(x) for x in tandem_frequency.keys()]):
|
|
|
227 o.write("{0}\t{1}\n".format(frq, tandem_frequency[str(frq)]))
|
|
0
|
228
|
|
47
|
229 tandem_row = []
|
|
|
230 genes_extra = list(genes)
|
|
|
231 genes_extra.append("all")
|
|
|
232 for x, y, in zip([tandem_sum_by_class[x] for x in genes_extra], [expected_tandem_sum_by_class[x] for x in genes_extra]):
|
|
|
233 if y != 0:
|
|
|
234 tandem_row += [x, round(y, 2), round(x / y, 2)]
|
|
|
235 else:
|
|
|
236 tandem_row += [x, round(y, 2), 0]
|
|
0
|
237
|
|
47
|
238 tandem_freq_file = os.path.join(os.path.dirname(outfile), "shm_overview_tandem_row.txt")
|
|
|
239 with open(tandem_freq_file, 'w') as o:
|
|
|
240 o.write("Tandems/Expected (ratio),{0}\n".format(",".join([str(x) for x in tandem_row])))
|
|
|
241
|
|
|
242 #print mutationList, linecount
|
|
|
243
|
|
62
|
244 AALength = (int(max(mutationList, key=lambda i: int(i[4]) if i[4] and i[5] != ";" else 0)[4]) + 1) # [4] is the position of the AA mutation, None if silent
|
|
47
|
245 if AALength < 60:
|
|
|
246 AALength = 64
|
|
|
247
|
|
|
248 AA_mutation = [0] * AALength
|
|
|
249 AA_mutation_dic = {"IGA": AA_mutation[:], "IGG": AA_mutation[:], "IGM": AA_mutation[:], "IGE": AA_mutation[:], "unm": AA_mutation[:], "all": AA_mutation[:]}
|
|
|
250 AA_mutation_empty = AA_mutation[:]
|
|
|
251
|
|
66
|
252 print "AALength:", AALength
|
|
47
|
253 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt"
|
|
|
254 with open(aa_mutations_by_id_file, 'w') as o:
|
|
|
255 o.write("ID\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
|
|
|
256 for ID in mutationListByID.keys():
|
|
|
257 AA_mutation_for_ID = AA_mutation_empty[:]
|
|
|
258 for mutation in mutationListByID[ID]:
|
|
66
|
259 if mutation[4] and mutation[5] != ";":
|
|
47
|
260 AA_mutation_position = int(mutation[4])
|
|
66
|
261 try:
|
|
|
262 AA_mutation[AA_mutation_position] += 1
|
|
|
263 AA_mutation_for_ID[AA_mutation_position] += 1
|
|
|
264 except Exception as e:
|
|
|
265 print e
|
|
|
266 print mutation
|
|
|
267 sys.exit()
|
|
47
|
268 clss = genedic[ID][:3]
|
|
|
269 AA_mutation_dic[clss][AA_mutation_position] += 1
|
|
|
270 o.write(ID + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
|
|
0
|
271
|
|
|
272
|
|
|
273
|
|
47
|
274 #absent AA stuff
|
|
|
275 absentAACDR1Dic = defaultdict(list)
|
|
|
276 absentAACDR1Dic[5] = range(29,36)
|
|
|
277 absentAACDR1Dic[6] = range(29,35)
|
|
|
278 absentAACDR1Dic[7] = range(30,35)
|
|
|
279 absentAACDR1Dic[8] = range(30,34)
|
|
|
280 absentAACDR1Dic[9] = range(31,34)
|
|
|
281 absentAACDR1Dic[10] = range(31,33)
|
|
|
282 absentAACDR1Dic[11] = [32]
|
|
0
|
283
|
|
47
|
284 absentAACDR2Dic = defaultdict(list)
|
|
|
285 absentAACDR2Dic[0] = range(55,65)
|
|
|
286 absentAACDR2Dic[1] = range(56,65)
|
|
|
287 absentAACDR2Dic[2] = range(56,64)
|
|
|
288 absentAACDR2Dic[3] = range(57,64)
|
|
|
289 absentAACDR2Dic[4] = range(57,63)
|
|
|
290 absentAACDR2Dic[5] = range(58,63)
|
|
|
291 absentAACDR2Dic[6] = range(58,62)
|
|
|
292 absentAACDR2Dic[7] = range(59,62)
|
|
|
293 absentAACDR2Dic[8] = range(59,61)
|
|
|
294 absentAACDR2Dic[9] = [60]
|
|
0
|
295
|
|
47
|
296 absentAA = [len(IDlist)] * (AALength-1)
|
|
|
297 for k, cdr1Length in cdr1LengthDic.iteritems():
|
|
|
298 for c in absentAACDR1Dic[cdr1Length]:
|
|
|
299 absentAA[c] -= 1
|
|
0
|
300
|
|
47
|
301 for k, cdr2Length in cdr2LengthDic.iteritems():
|
|
|
302 for c in absentAACDR2Dic[cdr2Length]:
|
|
|
303 absentAA[c] -= 1
|
|
0
|
304
|
|
|
305
|
|
47
|
306 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
|
|
|
307 with open(aa_mutations_by_id_file, 'w') as o:
|
|
|
308 o.write("ID\tcdr1length\tcdr2length\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
|
|
|
309 for ID in IDlist:
|
|
|
310 absentAAbyID = [1] * (AALength-1)
|
|
|
311 cdr1Length = cdr1LengthDic[ID]
|
|
|
312 for c in absentAACDR1Dic[cdr1Length]:
|
|
|
313 absentAAbyID[c] -= 1
|
|
0
|
314
|
|
47
|
315 cdr2Length = cdr2LengthDic[ID]
|
|
|
316 for c in absentAACDR2Dic[cdr2Length]:
|
|
|
317 absentAAbyID[c] -= 1
|
|
|
318 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
|
|
0
|
319
|
|
47
|
320 if linecount == 0:
|
|
|
321 print "No data, exiting"
|
|
|
322 with open(outfile, 'w') as o:
|
|
|
323 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
|
|
|
324 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
|
|
|
325 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
|
|
|
326 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
|
|
|
327 import sys
|
|
0
|
328
|
|
47
|
329 sys.exit()
|
|
0
|
330
|
|
47
|
331 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
|
|
|
332 RGYWCount = {}
|
|
|
333 WRCYCount = {}
|
|
|
334 WACount = {}
|
|
|
335 TWCount = {}
|
|
0
|
336
|
|
47
|
337 #IDIndex = 0
|
|
|
338 ataIndex = 0
|
|
|
339 tatIndex = 0
|
|
|
340 aggctatIndex = 0
|
|
|
341 atagcctIndex = 0
|
|
|
342 first = True
|
|
|
343 with open(infile, 'ru') as i:
|
|
|
344 for line in i:
|
|
|
345 if first:
|
|
|
346 linesplt = line.split("\t")
|
|
|
347 ataIndex = linesplt.index("X.a.t.a")
|
|
|
348 tatIndex = linesplt.index("t.a.t.")
|
|
|
349 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
|
|
|
350 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
|
|
|
351 first = False
|
|
|
352 continue
|
|
0
|
353 linesplt = line.split("\t")
|
|
47
|
354 gene = linesplt[best_matchIndex]
|
|
|
355 ID = linesplt[IDIndex]
|
|
|
356 RGYW = [(int(x), int(y), z) for (x, y, z) in
|
|
|
357 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
|
|
|
358 WRCY = [(int(x), int(y), z) for (x, y, z) in
|
|
|
359 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
|
|
|
360 WA = [(int(x), int(y), z) for (x, y, z) in
|
|
|
361 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
|
|
|
362 TW = [(int(x), int(y), z) for (x, y, z) in
|
|
|
363 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
|
|
|
364 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
|
|
0
|
365
|
|
62
|
366 with open(os.path.join(os.path.dirname(os.path.abspath(infile)), "RGYW.txt"), 'a') as out_handle:
|
|
|
367 for hotspot in RGYW:
|
|
|
368 out_handle.write("{0}\t{1}\n".format(ID, "\t".join([str(x) for x in hotspot])))
|
|
|
369
|
|
47
|
370 mutationList = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
|
|
|
371 for mutation in mutationList:
|
|
|
372 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
|
|
62
|
373 mutation_in_RGYW = any(((start <= int(where) <= end) for (start, end, region) in RGYW))
|
|
|
374 mutation_in_WRCY = any(((start <= int(where) <= end) for (start, end, region) in WRCY))
|
|
|
375 mutation_in_WA = any(((start <= int(where) <= end) for (start, end, region) in WA))
|
|
|
376 mutation_in_TW = any(((start <= int(where) <= end) for (start, end, region) in TW))
|
|
0
|
377
|
|
47
|
378 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
|
|
0
|
379
|
|
47
|
380 if in_how_many_motifs > 0:
|
|
|
381 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
|
|
|
382 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
|
|
|
383 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
|
|
|
384 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
|
|
63
|
385
|
|
|
386 mutations_in_motifs_file = os.path.join(os.path.dirname(os.path.abspath(infile)), "mutation_in_motifs.txt")
|
|
|
387 if not os.path.exists(mutation_by_id_file):
|
|
|
388 with open(mutations_in_motifs_file, 'w') as out_handle:
|
|
|
389 out_handle.write("{0}\n".format("\t".join([
|
|
|
390 "Sequence.ID",
|
|
|
391 "mutation_position",
|
|
|
392 "region",
|
|
|
393 "from_nt",
|
|
|
394 "to_nt",
|
|
|
395 "mutation_position_AA",
|
|
|
396 "from_AA",
|
|
|
397 "to_AA",
|
|
|
398 "motif",
|
|
|
399 "motif_start_nt",
|
|
|
400 "motif_end_nt",
|
|
|
401 "rest"
|
|
|
402 ])))
|
|
|
403
|
|
|
404 with open(mutations_in_motifs_file, 'a') as out_handle:
|
|
|
405 motif_dic = {"RGYW": RGYW, "WRCY": WRCY, "WA": WA, "TW": TW}
|
|
|
406 for mutation in mutationList:
|
|
|
407 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
|
|
|
408 for motif in motif_dic.keys():
|
|
|
409
|
|
|
410 for start, end, region in motif_dic[motif]:
|
|
|
411 if start <= int(where) <= end:
|
|
|
412 out_handle.write("{0}\n".format(
|
|
|
413 "\t".join([
|
|
|
414 ID,
|
|
|
415 where,
|
|
|
416 region,
|
|
|
417 frm,
|
|
|
418 to,
|
|
|
419 str(AAwhere),
|
|
|
420 str(AAfrm),
|
|
|
421 str(AAto),
|
|
|
422 motif,
|
|
|
423 str(start),
|
|
|
424 str(end),
|
|
|
425 str(junk)
|
|
|
426 ])
|
|
|
427 ))
|
|
|
428
|
|
0
|
429
|
|
|
430
|
|
47
|
431 def mean(lst):
|
|
|
432 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
|
|
0
|
433
|
|
|
434
|
|
47
|
435 def median(lst):
|
|
|
436 lst = sorted(lst)
|
|
|
437 l = len(lst)
|
|
|
438 if l == 0:
|
|
|
439 return 0
|
|
|
440 if l == 1:
|
|
|
441 return lst[0]
|
|
|
442
|
|
|
443 l = int(l / 2)
|
|
0
|
444
|
|
47
|
445 if len(lst) % 2 == 0:
|
|
|
446 return float(lst[l] + lst[(l - 1)]) / 2.0
|
|
|
447 else:
|
|
|
448 return lst[l]
|
|
0
|
449
|
|
47
|
450 funcs = {"mean": mean, "median": median, "sum": sum}
|
|
0
|
451
|
|
47
|
452 directory = outfile[:outfile.rfind("/") + 1]
|
|
|
453 value = 0
|
|
|
454 valuedic = dict()
|
|
0
|
455
|
|
47
|
456 for fname in funcs.keys():
|
|
|
457 for gene in genes:
|
|
|
458 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v:
|
|
|
459 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip())
|
|
|
460 with open(directory + "all_" + fname + "_value.txt", 'r') as v:
|
|
|
461 valuedic["total_" + fname] = float(v.readlines()[0].rstrip())
|
|
|
462
|
|
0
|
463
|
|
47
|
464 def get_xyz(lst, gene, f, fname):
|
|
|
465 x = round(round(f(lst), 1))
|
|
|
466 y = valuedic[gene + "_" + fname]
|
|
|
467 z = str(round(x / float(y) * 100, 1)) if y != 0 else "0"
|
|
|
468 return (str(x), str(y), z)
|
|
0
|
469
|
|
47
|
470 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
|
|
|
471 arr = ["RGYW", "WRCY", "WA", "TW"]
|
|
0
|
472
|
|
47
|
473 for fname in funcs.keys():
|
|
|
474 func = funcs[fname]
|
|
|
475 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt"
|
|
|
476 with open(foutfile, 'w') as o:
|
|
|
477 for typ in arr:
|
|
|
478 o.write(typ + " (%)")
|
|
|
479 curr = dic[typ]
|
|
|
480 for gene in genes:
|
|
|
481 geneMatcher = geneMatchers[gene]
|
|
|
482 if valuedic[gene + "_" + fname] is 0:
|
|
|
483 o.write(",0,0,0")
|
|
|
484 else:
|
|
|
485 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
|
|
|
486 o.write("," + x + "," + y + "," + z)
|
|
|
487 x, y, z = get_xyz([y for x, y in curr.iteritems() if not genedic[x].startswith("unmatched")], "total", func, fname)
|
|
|
488 #x, y, z = get_xyz([y for x, y in curr.iteritems()], "total", func, fname)
|
|
|
489 o.write("," + x + "," + y + "," + z + "\n")
|
|
0
|
490
|
|
|
491
|
|
47
|
492 # for testing
|
|
|
493 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
|
|
|
494 with open(seq_motif_file, 'w') as o:
|
|
|
495 o.write("ID\tRGYW\tWRCY\tWA\tTW\n")
|
|
|
496 for ID in IDlist:
|
|
|
497 #o.write(ID + "\t" + str(round(RGYWCount[ID], 2)) + "\t" + str(round(WRCYCount[ID], 2)) + "\t" + str(round(WACount[ID], 2)) + "\t" + str(round(TWCount[ID], 2)) + "\n")
|
|
|
498 o.write(ID + "\t" + str(RGYWCount[ID]) + "\t" + str(WRCYCount[ID]) + "\t" + str(WACount[ID]) + "\t" + str(TWCount[ID]) + "\n")
|
|
|
499
|
|
|
500 if __name__ == "__main__":
|
|
|
501 main()
|
