Mercurial > repos > davidvanzessen > shm_csr

comparison shm_csr.py @ 47:64711f461c8e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Thu, 04 May 2017 07:43:09 -0400
parents 77a7ac76c7b9
children c5295dd10dfc
comparison
equal deleted inserted replaced
46:cfc9a442e59d 47:64711f461c8e
1 from __future__ import division 1 import argparse
2 import logging
3 import sys
4 import os
5 import re
6
2 from collections import defaultdict 7 from collections import defaultdict
3 import re 8
4 import argparse 9 def main():
5 10 parser = argparse.ArgumentParser()
6 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")
7 parser.add_argument("--input", 12 parser.add_argument("--genes", help="The genes available in the 'best_match' column")
8 help="The '7_V-REGION-mutation-and-AA-change-table' and '10_V-REGION-mutation-hotspots' merged together, with an added 'best_match' annotation") 13 parser.add_argument("--empty_region_filter", help="Where does the sequence start?", choices=['leader', 'FR1', 'CDR1', 'FR2'])
9 parser.add_argument("--genes", help="The genes available in the 'best_match' column") 14 parser.add_argument("--output", help="Output file")
10 parser.add_argument("--empty_region_filter", help="Where does the sequence start?", choices=['leader', 'FR1', 'CDR1', 'FR2']) 15
11 parser.add_argument("--output", help="Output file") 16 args = parser.parse_args()
12 17
13 args = parser.parse_args() 18 infile = args.input
14 19 genes = str(args.genes).split(",")
15 infile = args.input 20 empty_region_filter = args.empty_region_filter
16 genes = str(args.genes).split(",") 21 outfile = args.output
17 empty_region_filter = args.empty_region_filter 22
18 outfile = args.output 23 genedic = dict()
19 24
20 genedic = dict() 25 mutationdic = dict()
21 26 mutationMatcher = re.compile("^(.)(\d+).(.),?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?")
22 mutationdic = dict() 27 NAMatchResult = (None, None, None, None, None, None, '')
23 mutationMatcher = re.compile("^(.)(\d+).(.),?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?") 28 geneMatchers = {gene: re.compile("^" + gene + ".*") for gene in genes}
24 NAMatchResult = (None, None, None, None, None, None, '') 29 linecount = 0
25 linecount = 0 30
26 31 IDIndex = 0
27 IDIndex = 0 32 best_matchIndex = 0
28 best_matchIndex = 0 33 fr1Index = 0
29 fr1Index = 0 34 cdr1Index = 0
30 cdr1Index = 0 35 fr2Index = 0
31 fr2Index = 0 36 cdr2Index = 0
32 cdr2Index = 0 37 fr3Index = 0
33 fr3Index = 0 38 first = True
34 first = True 39 IDlist = []
35 IDlist = [] 40 mutationList = []
36 mutationList = [] 41 mutationListByID = {}
37 mutationListByID = {} 42 cdr1LengthDic = {}
38 cdr1LengthDic = {} 43 cdr2LengthDic = {}
39 cdr2LengthDic = {} 44
40 45 fr1LengthDict = {}
41 with open(infile, 'ru') as i: 46 fr2LengthDict = {}
42 for line in i: 47 fr3LengthDict = {}
43 if first: 48
49 cdr1LengthIndex = 0
50 cdr2LengthIndex = 0
51
52 fr1SeqIndex = 0
53 fr2SeqIndex = 0
54 fr3SeqIndex = 0
55
56 tandem_sum_by_class = defaultdict(int)
57 expected_tandem_sum_by_class = defaultdict(float)
58
59 with open(infile, 'ru') as i:
60 for line in i:
61 if first:
62 linesplt = line.split("\t")
63 IDIndex = linesplt.index("Sequence.ID")
64 best_matchIndex = linesplt.index("best_match")
65 fr1Index = linesplt.index("FR1.IMGT")
66 cdr1Index = linesplt.index("CDR1.IMGT")
67 fr2Index = linesplt.index("FR2.IMGT")
68 cdr2Index = linesplt.index("CDR2.IMGT")
69 fr3Index = linesplt.index("FR3.IMGT")
70 cdr1LengthIndex = linesplt.index("CDR1.IMGT.seq")
71 cdr2LengthIndex = linesplt.index("CDR2.IMGT.seq")
72 fr1SeqIndex = linesplt.index("FR1.IMGT.seq")
73 fr2SeqIndex = linesplt.index("FR2.IMGT.seq")
74 fr3SeqIndex = linesplt.index("FR3.IMGT.seq")
75 first = False
76 continue
77 linecount += 1
44 linesplt = line.split("\t") 78 linesplt = line.split("\t")
45 IDIndex = linesplt.index("Sequence.ID") 79 ID = linesplt[IDIndex]
46 best_matchIndex = linesplt.index("best_match") 80 genedic[ID] = linesplt[best_matchIndex]
47 fr1Index = linesplt.index("FR1.IMGT") 81 try:
48 cdr1Index = linesplt.index("CDR1.IMGT") 82 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x] if (linesplt[fr1Index] != "NA" and empty_region_filter == "leader") else []
49 fr2Index = linesplt.index("FR2.IMGT") 83 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x] if (linesplt[cdr1Index] != "NA" and empty_region_filter in ["leader", "FR1"]) else []
50 cdr2Index = linesplt.index("CDR2.IMGT") 84 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x] if (linesplt[fr2Index] != "NA" and empty_region_filter in ["leader", "FR1", "CDR1"]) else []
51 fr3Index = linesplt.index("FR3.IMGT") 85 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x] if (linesplt[cdr2Index] != "NA") else []
52 cdr1LengthIndex = linesplt.index("CDR1.IMGT.length") 86 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
53 cdr2LengthIndex = linesplt.index("CDR2.IMGT.length") 87 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x] if linesplt[fr3Index] != "NA" else []
54 first = False 88 except Exception as e:
55 continue 89 print "Something went wrong while processing this line:"
56 linecount += 1 90 print linesplt
57 linesplt = line.split("\t") 91 print linecount
58 ID = linesplt[IDIndex] 92 print e
59 genedic[ID] = linesplt[best_matchIndex] 93 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
60 try: 94 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
61 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x] if (linesplt[fr1Index] != "NA" and empty_region_filter == "leader") else [] 95
62 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x] if (linesplt[cdr1Index] != "NA" and empty_region_filter in ["leader", "FR1"]) else [] 96 cdr1Length = len(linesplt[cdr1LengthIndex])
63 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x] if (linesplt[fr2Index] != "NA" and empty_region_filter in ["leader", "FR1", "CDR1"]) else [] 97 cdr2Length = len(linesplt[cdr2LengthIndex])
64 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x] if (linesplt[cdr2Index] != "NA") else [] 98
65 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] 99 #print linesplt[fr2SeqIndex]
66 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x] if linesplt[fr3Index] != "NA" else [] 100 fr1Length = len(linesplt[fr1SeqIndex]) if empty_region_filter == "leader" else 0
67 except Exception as e: 101 fr2Length = len(linesplt[fr2SeqIndex]) if empty_region_filter in ["leader", "FR1", "CDR1"] else 0
68 print "Something went wrong while processing this line:" 102 fr3Length = len(linesplt[fr3SeqIndex])
69 print linesplt 103
70 print linecount 104 cdr1LengthDic[ID] = cdr1Length
71 print e 105 cdr2LengthDic[ID] = cdr2Length
72 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"] 106
73 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"] 107 fr1LengthDict[ID] = fr1Length
74 108 fr2LengthDict[ID] = fr2Length
75 cdr1Length = linesplt[cdr1LengthIndex] 109 fr3LengthDict[ID] = fr3Length
76 cdr2Length = linesplt[cdr2LengthIndex] 110
77 111 IDlist += [ID]
78 cdr1LengthDic[ID] = int(cdr1Length) if cdr1Length != "X" else 0 112
79 cdr2LengthDic[ID] = int(cdr2Length) if cdr2Length != "X" else 0 113
114 #tandem mutation stuff
115 tandem_frequency = defaultdict(int)
116 mutation_frequency = defaultdict(int)
117
118 tandem_file = os.path.join(os.path.dirname(outfile), "tandems_by_id.txt")
119 with open(tandem_file, 'w') as o:
120 highest_tandem_length = 0
121
122 o.write("Sequence.ID\tnumber_of_mutations\tnumber_of_tandems\tregion_length\texpected_tandems\tlongest_tandem\ttandems\n")
123 for ID in IDlist:
124 mutations = mutationListByID[ID]
125 if len(mutations) == 0:
126 continue
127 last_mut = max(mutations, key=lambda x: int(x[1]))
128
129 last_mut_pos = int(last_mut[1])
130
131 mut_positions = [False] * (last_mut_pos + 1)
132
133 for mutation in mutations:
134 frm, where, to, frmAA, whereAA, toAA, thing = mutation
135 where = int(where)
136 mut_positions[where] = True
137
138 tandem_muts = []
139 tandem_start = -1
140 tandem_length = 0
141 for i in range(len(mut_positions)):
142 if mut_positions[i]:
143 if tandem_start == -1:
144 tandem_start = i
145 tandem_length += 1
146 #print "".join(["1" if x else "0" for x in mut_positions[:i+1]])
147 else:
148 if tandem_length > 1:
149 tandem_muts.append((tandem_start, tandem_length))
150 #print "{0}{1} {2}:{3}".format(" " * (i - tandem_length), "^" * tandem_length, tandem_start, tandem_length)
151 tandem_start = -1
152 tandem_length = 0
153 if tandem_length > 1: # if the sequence ends with a tandem mutation
154 tandem_muts.append((tandem_start, tandem_length))
155
156 if len(tandem_muts) > 0:
157 if highest_tandem_length < len(tandem_muts):
158 highest_tandem_length = len(tandem_muts)
159
160 region_length = fr1LengthDict[ID] + cdr1LengthDic[ID] + fr2LengthDict[ID] + cdr2LengthDic[ID] + fr3LengthDict[ID]
161 longest_tandem = max(tandem_muts, key=lambda x: x[1]) if len(tandem_muts) else (0, 0)
162 num_mutations = len(mutations)
163 f_num_mutations = float(num_mutations)
164 num_tandem_muts = len(tandem_muts)
165 expected_tandem_muts = f_num_mutations * (f_num_mutations - 1.0) / float(region_length)
166 o.write("{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\n".format(ID,
167 str(num_mutations),
168 str(num_tandem_muts),
169 str(region_length),
170 str(round(expected_tandem_muts, 2)),
171 str(longest_tandem[1]),
172 str(tandem_muts)))
173 gene = genedic[ID]
174 if gene.find("unmatched") == -1:
175 tandem_sum_by_class[gene] += num_tandem_muts
176 expected_tandem_sum_by_class[gene] += expected_tandem_muts
177
178 tandem_sum_by_class["all"] += num_tandem_muts
179 expected_tandem_sum_by_class["all"] += expected_tandem_muts
180
181 gene = gene[:3]
182 if gene in ["IGA", "IGG"]:
183 tandem_sum_by_class[gene] += num_tandem_muts
184 expected_tandem_sum_by_class[gene] += expected_tandem_muts
185 else:
186 tandem_sum_by_class["unmatched"] += num_tandem_muts
187 expected_tandem_sum_by_class["unmatched"] += expected_tandem_muts
188
189
190 for tandem_mut in tandem_muts:
191 tandem_frequency[str(tandem_mut[1])] += 1
192 #print "\t".join([ID, str(len(tandem_muts)), str(longest_tandem[1]) , str(tandem_muts)])
193
194 tandem_freq_file = os.path.join(os.path.dirname(outfile), "tandem_frequency.txt")
195 with open(tandem_freq_file, 'w') as o:
196 for frq in sorted([int(x) for x in tandem_frequency.keys()]):
197 o.write("{0}\t{1}\n".format(frq, tandem_frequency[str(frq)]))
198
199 tandem_row = []
200 print genes
201 print tandem_sum_by_class
202 print expected_tandem_sum_by_class
203 genes_extra = list(genes)
204 genes_extra.append("all")
205 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]):
206 if y != 0:
207 tandem_row += [x, round(y, 2), round(x / y, 2)]
208 else:
209 tandem_row += [x, round(y, 2), 0]
210
211 """
212 print tandem_row
213 tandem_row += tandem_row[-3:]
214 print tandem_row
215 all_expected_tandem = expected_tandem_sum_by_class["all"]
216 all_tandem = tandem_sum_by_class["all"]
217 if all_expected_tandem == 0:
218 tandem_row[-6:-3] = [all_tandem, round(all_expected_tandem, 2), 0]
219 else:
220 tandem_row[-6:-3] = [all_tandem, round(all_expected_tandem, 2), round(all_tandem / all_expected_tandem, 2)]
221 print tandem_row
222 """
223 for i in range(len(genes_extra)):
224 gene = genes_extra[i]
225 print gene, tandem_row[i*3:i*3+3]
226
227 tandem_freq_file = os.path.join(os.path.dirname(outfile), "shm_overview_tandem_row.txt")
228 with open(tandem_freq_file, 'w') as o:
229 o.write("Tandems/Expected (ratio),{0}\n".format(",".join([str(x) for x in tandem_row])))
230
231 #print mutationList, linecount
232
233 AALength = (int(max(mutationList, key=lambda i: int(i[4]) if i[4] else 0)[4]) + 1) # [4] is the position of the AA mutation, None if silent
234 if AALength < 60:
235 AALength = 64
236
237 AA_mutation = [0] * AALength
238 AA_mutation_dic = {"IGA": AA_mutation[:], "IGG": AA_mutation[:], "IGM": AA_mutation[:], "IGE": AA_mutation[:], "unm": AA_mutation[:], "all": AA_mutation[:]}
239 AA_mutation_empty = AA_mutation[:]
240
241 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt"
242 with open(aa_mutations_by_id_file, 'w') as o:
243 o.write("ID\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
244 for ID in mutationListByID.keys():
245 AA_mutation_for_ID = AA_mutation_empty[:]
246 for mutation in mutationListByID[ID]:
247 if mutation[4]:
248 AA_mutation_position = int(mutation[4])
249 AA_mutation[AA_mutation_position] += 1
250 AA_mutation_for_ID[AA_mutation_position] += 1
251 clss = genedic[ID][:3]
252 AA_mutation_dic[clss][AA_mutation_position] += 1
253 o.write(ID + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
254
255
256
257 #absent AA stuff
258 absentAACDR1Dic = defaultdict(list)
259 absentAACDR1Dic[5] = range(29,36)
260 absentAACDR1Dic[6] = range(29,35)
261 absentAACDR1Dic[7] = range(30,35)
262 absentAACDR1Dic[8] = range(30,34)
263 absentAACDR1Dic[9] = range(31,34)
264 absentAACDR1Dic[10] = range(31,33)
265 absentAACDR1Dic[11] = [32]
266
267 absentAACDR2Dic = defaultdict(list)
268 absentAACDR2Dic[0] = range(55,65)
269 absentAACDR2Dic[1] = range(56,65)
270 absentAACDR2Dic[2] = range(56,64)
271 absentAACDR2Dic[3] = range(57,64)
272 absentAACDR2Dic[4] = range(57,63)
273 absentAACDR2Dic[5] = range(58,63)
274 absentAACDR2Dic[6] = range(58,62)
275 absentAACDR2Dic[7] = range(59,62)
276 absentAACDR2Dic[8] = range(59,61)
277 absentAACDR2Dic[9] = [60]
278
279 absentAA = [len(IDlist)] * (AALength-1)
280 for k, cdr1Length in cdr1LengthDic.iteritems():
281 for c in absentAACDR1Dic[cdr1Length]:
282 absentAA[c] -= 1
283
284 for k, cdr2Length in cdr2LengthDic.iteritems():
285 for c in absentAACDR2Dic[cdr2Length]:
286 absentAA[c] -= 1
287
288
289 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
290 with open(aa_mutations_by_id_file, 'w') as o:
291 o.write("ID\tcdr1length\tcdr2length\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
292 for ID in IDlist:
293 absentAAbyID = [1] * (AALength-1)
294 cdr1Length = cdr1LengthDic[ID]
295 for c in absentAACDR1Dic[cdr1Length]:
296 absentAAbyID[c] -= 1
297
298 cdr2Length = cdr2LengthDic[ID]
299 for c in absentAACDR2Dic[cdr2Length]:
300 absentAAbyID[c] -= 1
301 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
302
303 if linecount == 0:
304 print "No data, exiting"
305 with open(outfile, 'w') as o:
306 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
307 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
308 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
309 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
310 import sys
311
312 sys.exit()
313
314 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
315 RGYWCount = {}
316 WRCYCount = {}
317 WACount = {}
318 TWCount = {}
319
320 #IDIndex = 0
321 ataIndex = 0
322 tatIndex = 0
323 aggctatIndex = 0
324 atagcctIndex = 0
325 first = True
326 with open(infile, 'ru') as i:
327 for line in i:
328 if first:
329 linesplt = line.split("\t")
330 ataIndex = linesplt.index("X.a.t.a")
331 tatIndex = linesplt.index("t.a.t.")
332 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
333 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
334 first = False
335 continue
336 linesplt = line.split("\t")
337 gene = linesplt[best_matchIndex]
338 ID = linesplt[IDIndex]
339 RGYW = [(int(x), int(y), z) for (x, y, z) in
340 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
341 WRCY = [(int(x), int(y), z) for (x, y, z) in
342 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
343 WA = [(int(x), int(y), z) for (x, y, z) in
344 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
345 TW = [(int(x), int(y), z) for (x, y, z) in
346 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
347 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
348
349 mutationList = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
350 for mutation in mutationList:
351 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
352 mutation_in_RGYW = any([(start <= int(where) <= end) for (start, end, region) in RGYW])
353 mutation_in_WRCY = any([(start <= int(where) <= end) for (start, end, region) in WRCY])
354 mutation_in_WA = any([(start <= int(where) <= end) for (start, end, region) in WA])
355 mutation_in_TW = any([(start <= int(where) <= end) for (start, end, region) in TW])
356
357 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
358
359 if in_how_many_motifs > 0:
360 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
361 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
362 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
363 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
364
365
366 def mean(lst):
367 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
368
369
370 def median(lst):
371 lst = sorted(lst)
372 l = len(lst)
373 if l == 0:
374 return 0
375 if l == 1:
376 return lst[0]
377
378 l = int(l / 2)
80 379
81 IDlist += [ID] 380 if len(lst) % 2 == 0:
82 381 return float(lst[l] + lst[(l - 1)]) / 2.0
83 #print mutationList, linecount 382 else:
84 383 return lst[l]
85 AALength = (int(max(mutationList, key=lambda i: int(i[4]) if i[4] else 0)[4]) + 1) # [4] is the position of the AA mutation, None if silent 384
86 if AALength < 60: 385 funcs = {"mean": mean, "median": median, "sum": sum}
87 AALength = 64 386
88 387 directory = outfile[:outfile.rfind("/") + 1]
89 AA_mutation = [0] * AALength 388 value = 0
90 AA_mutation_dic = {"IGA": AA_mutation[:], "IGG": AA_mutation[:], "IGM": AA_mutation[:], "IGE": AA_mutation[:], "unm": AA_mutation[:], "all": AA_mutation[:]} 389 valuedic = dict()
91 AA_mutation_empty = AA_mutation[:] 390
92 391 for fname in funcs.keys():
93 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt" 392 for gene in genes:
94 with open(aa_mutations_by_id_file, 'w') as o: 393 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v:
95 o.write("ID\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n") 394 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip())
96 for ID in mutationListByID.keys(): 395 with open(directory + "all_" + fname + "_value.txt", 'r') as v:
97 AA_mutation_for_ID = AA_mutation_empty[:] 396 valuedic["total_" + fname] = float(v.readlines()[0].rstrip())
98 for mutation in mutationListByID[ID]:
99 if mutation[4]:
100 AA_mutation_position = int(mutation[4])
101 AA_mutation[AA_mutation_position] += 1
102 AA_mutation_for_ID[AA_mutation_position] += 1
103 clss = genedic[ID][:3]
104 AA_mutation_dic[clss][AA_mutation_position] += 1
105 o.write(ID + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
106
107
108
109 #absent AA stuff
110 absentAACDR1Dic = defaultdict(list)
111 absentAACDR1Dic[5] = range(29,36)
112 absentAACDR1Dic[6] = range(29,35)
113 absentAACDR1Dic[7] = range(30,35)
114 absentAACDR1Dic[8] = range(30,34)
115 absentAACDR1Dic[9] = range(31,34)
116 absentAACDR1Dic[10] = range(31,33)
117 absentAACDR1Dic[11] = [32]
118
119 absentAACDR2Dic = defaultdict(list)
120 absentAACDR2Dic[0] = range(55,65)
121 absentAACDR2Dic[1] = range(56,65)
122 absentAACDR2Dic[2] = range(56,64)
123 absentAACDR2Dic[3] = range(57,64)
124 absentAACDR2Dic[4] = range(57,63)
125 absentAACDR2Dic[5] = range(58,63)
126 absentAACDR2Dic[6] = range(58,62)
127 absentAACDR2Dic[7] = range(59,62)
128 absentAACDR2Dic[8] = range(59,61)
129 absentAACDR2Dic[9] = [60]
130
131 absentAA = [len(IDlist)] * (AALength-1)
132 for k, cdr1Length in cdr1LengthDic.iteritems():
133 for c in absentAACDR1Dic[cdr1Length]:
134 absentAA[c] -= 1
135
136 for k, cdr2Length in cdr2LengthDic.iteritems():
137 for c in absentAACDR2Dic[cdr2Length]:
138 absentAA[c] -= 1
139
140
141 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
142 with open(aa_mutations_by_id_file, 'w') as o:
143 o.write("ID\tcdr1length\tcdr2length\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
144 for ID in IDlist:
145 absentAAbyID = [1] * (AALength-1)
146 cdr1Length = cdr1LengthDic[ID]
147 for c in absentAACDR1Dic[cdr1Length]:
148 absentAAbyID[c] -= 1
149
150 cdr2Length = cdr2LengthDic[ID]
151 for c in absentAACDR2Dic[cdr2Length]:
152 absentAAbyID[c] -= 1
153 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
154
155 if linecount == 0:
156 print "No data, exiting"
157 with open(outfile, 'w') as o:
158 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
159 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
160 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
161 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
162 import sys
163
164 sys.exit()
165
166 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
167 RGYWCount = {}
168 WRCYCount = {}
169 WACount = {}
170 TWCount = {}
171
172 #IDIndex = 0
173 ataIndex = 0
174 tatIndex = 0
175 aggctatIndex = 0
176 atagcctIndex = 0
177 first = True
178 with open(infile, 'ru') as i:
179 for line in i:
180 if first:
181 linesplt = line.split("\t")
182 ataIndex = linesplt.index("X.a.t.a")
183 tatIndex = linesplt.index("t.a.t.")
184 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
185 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
186 first = False
187 continue
188 linesplt = line.split("\t")
189 gene = linesplt[best_matchIndex]
190 ID = linesplt[IDIndex]
191 RGYW = [(int(x), int(y), z) for (x, y, z) in
192 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
193 WRCY = [(int(x), int(y), z) for (x, y, z) in
194 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
195 WA = [(int(x), int(y), z) for (x, y, z) in
196 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
197 TW = [(int(x), int(y), z) for (x, y, z) in
198 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
199 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
200
201 mutationList = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
202 for mutation in mutationList:
203 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
204 mutation_in_RGYW = any([(start <= int(where) <= end) for (start, end, region) in RGYW])
205 mutation_in_WRCY = any([(start <= int(where) <= end) for (start, end, region) in WRCY])
206 mutation_in_WA = any([(start <= int(where) <= end) for (start, end, region) in WA])
207 mutation_in_TW = any([(start <= int(where) <= end) for (start, end, region) in TW])
208
209 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
210
211 if in_how_many_motifs > 0:
212 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
213 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
214 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
215 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
216
217
218 def mean(lst):
219 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
220
221
222 def median(lst):
223 lst = sorted(lst)
224 l = len(lst)
225 if l == 0:
226 return 0
227 if l == 1:
228 return lst[0]
229 397
230 l = int(l / 2) 398
231 399 def get_xyz(lst, gene, f, fname):
232 if len(lst) % 2 == 0: 400 x = round(round(f(lst), 1))
233 return float(lst[l] + lst[(l - 1)]) / 2.0 401 y = valuedic[gene + "_" + fname]
234 else: 402 z = str(round(x / float(y) * 100, 1)) if y != 0 else "0"
235 return lst[l] 403 return (str(x), str(y), z)
236 404
237 funcs = {"mean": mean, "median": median, "sum": sum} 405 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
238 406 arr = ["RGYW", "WRCY", "WA", "TW"]
239 directory = outfile[:outfile.rfind("/") + 1] 407
240 value = 0 408 for fname in funcs.keys():
241 valuedic = dict() 409 func = funcs[fname]
242 410 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt"
243 for fname in funcs.keys(): 411 with open(foutfile, 'w') as o:
244 for gene in genes: 412 for typ in arr:
245 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v: 413 o.write(typ + " (%)")
246 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip()) 414 curr = dic[typ]
247 with open(directory + "all_" + fname + "_value.txt", 'r') as v: 415 for gene in genes:
248 valuedic["total_" + fname] = float(v.readlines()[0].rstrip()) 416 geneMatcher = geneMatchers[gene]
249 417 if valuedic[gene + "_" + fname] is 0:
250 418 o.write(",0,0,0")
251 def get_xyz(lst, gene, f, fname): 419 else:
252 x = round(round(f(lst), 1)) 420 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
253 y = valuedic[gene + "_" + fname] 421 o.write("," + x + "," + y + "," + z)
254 z = str(round(x / float(y) * 100, 1)) if y != 0 else "0" 422 x, y, z = get_xyz([y for x, y in curr.iteritems() if not genedic[x].startswith("unmatched")], "total", func, fname)
255 return (str(x), str(y), z) 423 #x, y, z = get_xyz([y for x, y in curr.iteritems()], "total", func, fname)
256 424 o.write("," + x + "," + y + "," + z + "\n")
257 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount} 425
258 arr = ["RGYW", "WRCY", "WA", "TW"] 426
259 427 # for testing
260 geneMatchers = {gene: re.compile("^" + gene + ".*") for gene in genes} 428 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
261 429 with open(seq_motif_file, 'w') as o:
262 for fname in funcs.keys(): 430 o.write("ID\tRGYW\tWRCY\tWA\tTW\n")
263 func = funcs[fname] 431 for ID in IDlist:
264 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt" 432 #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")
265 with open(foutfile, 'w') as o: 433 o.write(ID + "\t" + str(RGYWCount[ID]) + "\t" + str(WRCYCount[ID]) + "\t" + str(WACount[ID]) + "\t" + str(TWCount[ID]) + "\n")
266 for typ in arr: 434
267 o.write(typ + " (%)") 435 if __name__ == "__main__":
268 curr = dic[typ] 436 main()
269 for gene in genes:
270 geneMatcher = geneMatchers[gene]
271 if valuedic[gene + "_" + fname] is 0:
272 o.write(",0,0,0")
273 else:
274 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
275 o.write("," + x + "," + y + "," + z)
276 x, y, z = get_xyz([y for x, y in curr.iteritems() if not genedic[x].startswith("unmatched")], "total", func, fname)
277 #x, y, z = get_xyz([y for x, y in curr.iteritems()], "total", func, fname)
278 o.write("," + x + "," + y + "," + z + "\n")
279
280
281 # for testing
282 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
283 with open(seq_motif_file, 'w') as o:
284 o.write("ID\tRGYW\tWRCY\tWA\tTW\n")
285 for ID in IDlist:
286 #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")
287 o.write(ID + "\t" + str(RGYWCount[ID]) + "\t" + str(WRCYCount[ID]) + "\t" + str(WACount[ID]) + "\t" + str(TWCount[ID]) + "\n")