Mercurial > repos > davidvanzessen > argalaxy_tools

comparison shm_csr.py @ 4:5ffd52fc35c4 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Mon, 12 Dec 2016 05:22:37 -0500
parents
children
comparison
equal deleted inserted replaced
3:beaa487ecf43 4:5ffd52fc35c4
1 from __future__ import division
2 from collections import defaultdict
3 import re
4 import argparse
5
6 parser = argparse.ArgumentParser()
7 parser.add_argument("--input",
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")
9 parser.add_argument("--genes", help="The genes available in the 'best_match' column")
10 parser.add_argument("--empty_region_filter", help="Where does the sequence start?", choices=['leader', 'FR1', 'CDR1', 'FR2'])
11 parser.add_argument("--output", help="Output file")
12
13 args = parser.parse_args()
14
15 infile = args.input
16 genes = str(args.genes).split(",")
17 empty_region_filter = args.empty_region_filter
18 outfile = args.output
19
20 genedic = dict()
21
22 mutationdic = dict()
23 mutationMatcher = re.compile("^(.)(\d+).(.),?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?")
24 NAMatchResult = (None, None, None, None, None, None, '')
25 linecount = 0
26
27 IDIndex = 0
28 best_matchIndex = 0
29 fr1Index = 0
30 cdr1Index = 0
31 fr2Index = 0
32 cdr2Index = 0
33 fr3Index = 0
34 first = True
35 IDlist = []
36 mutationList = []
37 mutationListByID = {}
38 cdr1LengthDic = {}
39 cdr2LengthDic = {}
40
41 with open(infile, 'r') as i:
42 for line in i:
43 if first:
44 linesplt = line.split("\t")
45 IDIndex = linesplt.index("Sequence.ID")
46 best_matchIndex = linesplt.index("best_match")
47 fr1Index = linesplt.index("FR1.IMGT")
48 cdr1Index = linesplt.index("CDR1.IMGT")
49 fr2Index = linesplt.index("FR2.IMGT")
50 cdr2Index = linesplt.index("CDR2.IMGT")
51 fr3Index = linesplt.index("FR3.IMGT")
52 cdr1LengthIndex = linesplt.index("CDR1.IMGT.length")
53 cdr2LengthIndex = linesplt.index("CDR2.IMGT.length")
54 first = False
55 continue
56 linecount += 1
57 linesplt = line.split("\t")
58 ID = linesplt[IDIndex]
59 genedic[ID] = linesplt[best_matchIndex]
60 try:
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 []
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 []
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 []
64 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x] if (linesplt[cdr2Index] != "NA") else []
65 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
66 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x] if linesplt[fr3Index] != "NA" else []
67 except Exception as e:
68 print "Something went wrong while processing this line:"
69 print linesplt
70 print linecount
71 print e
72 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
73 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
74
75 cdr1Length = linesplt[cdr1LengthIndex]
76 cdr2Length = linesplt[cdr2LengthIndex]
77
78 cdr1LengthDic[ID] = int(cdr1Length) if cdr1Length != "X" else 0
79 cdr2LengthDic[ID] = int(cdr2Length) if cdr2Length != "X" else 0
80
81 IDlist += [ID]
82
83 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
84 if AALength < 60:
85 AALength = 64
86
87 AA_mutation = [0] * AALength
88 AA_mutation_dic = {"IGA": AA_mutation[:], "IGG": AA_mutation[:], "IGM": AA_mutation[:], "IGE": AA_mutation[:], "unm": AA_mutation[:], "all": AA_mutation[:]}
89 AA_mutation_empty = AA_mutation[:]
90
91 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt"
92 with open(aa_mutations_by_id_file, 'w') as o:
93 o.write("ID\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
94 for ID in mutationListByID.keys():
95 AA_mutation_for_ID = AA_mutation_empty[:]
96 for mutation in mutationListByID[ID]:
97 if mutation[4]:
98 AA_mutation_position = int(mutation[4])
99 AA_mutation[AA_mutation_position] += 1
100 AA_mutation_for_ID[AA_mutation_position] += 1
101 clss = genedic[ID][:3]
102 AA_mutation_dic[clss][AA_mutation_position] += 1
103 o.write(ID + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
104
105
106
107 #absent AA stuff
108 absentAACDR1Dic = defaultdict(list)
109 absentAACDR1Dic[5] = range(29,36)
110 absentAACDR1Dic[6] = range(29,35)
111 absentAACDR1Dic[7] = range(30,35)
112 absentAACDR1Dic[8] = range(30,34)
113 absentAACDR1Dic[9] = range(31,34)
114 absentAACDR1Dic[10] = range(31,33)
115 absentAACDR1Dic[11] = [32]
116
117 absentAACDR2Dic = defaultdict(list)
118 absentAACDR2Dic[0] = range(55,65)
119 absentAACDR2Dic[1] = range(56,65)
120 absentAACDR2Dic[2] = range(56,64)
121 absentAACDR2Dic[3] = range(57,64)
122 absentAACDR2Dic[4] = range(57,63)
123 absentAACDR2Dic[5] = range(58,63)
124 absentAACDR2Dic[6] = range(58,62)
125 absentAACDR2Dic[7] = range(59,62)
126 absentAACDR2Dic[8] = range(59,61)
127 absentAACDR2Dic[9] = [60]
128
129 absentAA = [len(IDlist)] * (AALength-1)
130 for k, cdr1Length in cdr1LengthDic.iteritems():
131 for c in absentAACDR1Dic[cdr1Length]:
132 absentAA[c] -= 1
133
134 for k, cdr2Length in cdr2LengthDic.iteritems():
135 for c in absentAACDR2Dic[cdr2Length]:
136 absentAA[c] -= 1
137
138
139 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
140 with open(aa_mutations_by_id_file, 'w') as o:
141 o.write("ID\tcdr1length\tcdr2length\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
142 for ID in IDlist:
143 absentAAbyID = [1] * (AALength-1)
144 cdr1Length = cdr1LengthDic[ID]
145 for c in absentAACDR1Dic[cdr1Length]:
146 absentAAbyID[c] -= 1
147
148 cdr2Length = cdr2LengthDic[ID]
149 for c in absentAACDR2Dic[cdr2Length]:
150 absentAAbyID[c] -= 1
151 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
152
153 if linecount == 0:
154 print "No data, exiting"
155 with open(outfile, 'w') as o:
156 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
157 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
158 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
159 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
160 import sys
161
162 sys.exit()
163
164 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
165 RGYWCount = {}
166 WRCYCount = {}
167 WACount = {}
168 TWCount = {}
169
170 #IDIndex = 0
171 ataIndex = 0
172 tatIndex = 0
173 aggctatIndex = 0
174 atagcctIndex = 0
175 first = True
176 with open(infile, 'r') as i:
177 for line in i:
178 if first:
179 linesplt = line.split("\t")
180 ataIndex = linesplt.index("X.a.t.a")
181 tatIndex = linesplt.index("t.a.t.")
182 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
183 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
184 first = False
185 continue
186 linesplt = line.split("\t")
187 gene = linesplt[best_matchIndex]
188 ID = linesplt[IDIndex]
189 RGYW = [(int(x), int(y), z) for (x, y, z) in
190 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
191 WRCY = [(int(x), int(y), z) for (x, y, z) in
192 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
193 WA = [(int(x), int(y), z) for (x, y, z) in
194 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
195 TW = [(int(x), int(y), z) for (x, y, z) in
196 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
197 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
198
199 mutationList = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
200 for mutation in mutationList:
201 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
202 mutation_in_RGYW = any([(start <= int(where) <= end) for (start, end, region) in RGYW])
203 mutation_in_WRCY = any([(start <= int(where) <= end) for (start, end, region) in WRCY])
204 mutation_in_WA = any([(start <= int(where) <= end) for (start, end, region) in WA])
205 mutation_in_TW = any([(start <= int(where) <= end) for (start, end, region) in TW])
206
207 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
208
209 if in_how_many_motifs > 0:
210 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
211 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
212 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
213 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
214
215
216 def mean(lst):
217 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
218
219
220 def median(lst):
221 lst = sorted(lst)
222 l = len(lst)
223 if l == 0:
224 return 0
225 if l == 1:
226 return lst[0]
227
228 l = int(l / 2)
229
230 if len(lst) % 2 == 0:
231 return float(lst[l] + lst[(l - 1)]) / 2.0
232 else:
233 return lst[l]
234
235 funcs = {"mean": mean, "median": median, "sum": sum}
236
237 directory = outfile[:outfile.rfind("/") + 1]
238 value = 0
239 valuedic = dict()
240
241 for fname in funcs.keys():
242 for gene in genes:
243 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v:
244 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip())
245 with open(directory + "all_" + fname + "_value.txt", 'r') as v:
246 valuedic["total_" + fname] = float(v.readlines()[0].rstrip())
247
248
249 def get_xyz(lst, gene, f, fname):
250 x = round(round(f(lst), 1))
251 y = valuedic[gene + "_" + fname]
252 z = str(round(x / float(y) * 100, 1)) if y != 0 else "0"
253 return (str(x), str(y), z)
254
255 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
256 arr = ["RGYW", "WRCY", "WA", "TW"]
257
258 geneMatchers = {gene: re.compile("^" + gene + ".*") for gene in genes}
259
260 for fname in funcs.keys():
261 func = funcs[fname]
262 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt"
263 with open(foutfile, 'w') as o:
264 for typ in arr:
265 o.write(typ + " (%)")
266 curr = dic[typ]
267 for gene in genes:
268 geneMatcher = geneMatchers[gene]
269 if valuedic[gene + "_" + fname] is 0:
270 o.write(",0,0,0")
271 else:
272 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
273 o.write("," + x + "," + y + "," + z)
274 x, y, z = get_xyz([y for x, y in curr.iteritems() if not genedic[x].startswith("unmatched")], "total", func, fname)
275 #x, y, z = get_xyz([y for x, y in curr.iteritems()], "total", func, fname)
276 o.write("," + x + "," + y + "," + z + "\n")
277
278
279 # for testing
280 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
281 with open(seq_motif_file, 'w') as o:
282 o.write("ID\tRGYW\tWRCY\tWA\tTW\n")
283 for ID in IDlist:
284 #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")
285 o.write(ID + "\t" + str(RGYWCount[ID]) + "\t" + str(WRCYCount[ID]) + "\t" + str(WACount[ID]) + "\t" + str(TWCount[ID]) + "\n")