Mercurial > repos > davidvanzessen > argalaxy_tools

comparison mutation_analysis.py @ 57:16c7fc1c4bf8 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Fri, 18 Mar 2016 07:50:34 -0400
parents
children
comparison
equal deleted inserted replaced
56:2eb94c08e550 57:16c7fc1c4bf8
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("--includefr1", help="The genes available in the 'best_match' column")
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 print "includefr1 =", args.includefr1
18 include_fr1 = True if args.includefr1 == "yes" else False
19 outfile = args.output
20
21 genedic = dict()
22
23 mutationdic = dict()
24 mutationMatcher = re.compile("^(.)(\d+).(.),?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?")
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.Nb.of.nucleotides")
53 cdr2LengthIndex = linesplt.index("CDR2.IMGT.Nb.of.nucleotides")
54 first = False
55 continue
56 linecount += 1
57 linesplt = line.split("\t")
58 ID = linesplt[IDIndex]
59 genedic[ID] = linesplt[best_matchIndex]
60 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x] if include_fr1 else []
61 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x]
62 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x]
63 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x]
64 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
65 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x]
66
67 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
68 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
69
70 cdr1Length = linesplt[cdr1LengthIndex]
71 cdr2Length = linesplt[cdr2LengthIndex]
72
73 cdr1LengthDic[ID] = int(cdr1Length) / 3
74 cdr2LengthDic[ID] = int(cdr2Length) / 3
75
76 IDlist += [ID]
77
78 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
79
80 AA_mutation = [0] * AALength
81 AA_mutation_empty = AA_mutation[:]
82
83 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt"
84 with open(aa_mutations_by_id_file, 'w') as o:
85 for ID in mutationListByID.keys():
86 AA_mutation_for_ID = AA_mutation_empty[:]
87 for mutation in mutationListByID[ID]:
88 if mutation[4]:
89 AA_mutation[int(mutation[4])] += 1
90 AA_mutation_for_ID[int(mutation[4])] += 1
91 o.write(ID + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
92
93
94
95 #absent AA stuff
96 absentAACDR1Dic = defaultdict(list)
97 absentAACDR1Dic[5] = range(29,36)
98 absentAACDR1Dic[6] = range(29,35)
99 absentAACDR1Dic[7] = range(30,35)
100 absentAACDR1Dic[8] = range(30,34)
101 absentAACDR1Dic[9] = range(31,34)
102 absentAACDR1Dic[10] = range(31,33)
103 absentAACDR1Dic[11] = [32]
104
105 absentAACDR2Dic = defaultdict(list)
106 absentAACDR2Dic[0] = range(55,65)
107 absentAACDR2Dic[1] = range(56,65)
108 absentAACDR2Dic[2] = range(56,64)
109 absentAACDR2Dic[3] = range(57,64)
110 absentAACDR2Dic[4] = range(57,63)
111 absentAACDR2Dic[5] = range(58,63)
112 absentAACDR2Dic[6] = range(58,62)
113 absentAACDR2Dic[7] = range(59,62)
114 absentAACDR2Dic[8] = range(59,61)
115 absentAACDR2Dic[9] = [60]
116
117 absentAA = [len(IDlist)] * (AALength-1)
118 for k, cdr1Length in cdr1LengthDic.iteritems():
119 for c in absentAACDR1Dic[cdr1Length]:
120 absentAA[c] -= 1
121
122 for k, cdr2Length in cdr2LengthDic.iteritems():
123 for c in absentAACDR2Dic[cdr2Length]:
124 absentAA[c] -= 1
125
126
127 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
128 with open(aa_mutations_by_id_file, 'w') as o:
129 o.write("ID\tcdr1length\tcdr2length\t" + "\t".join([str(x) for x in range(1,AALength-1)]) + "\n")
130 for ID in IDlist:
131 absentAAbyID = [1] * (AALength-1)
132 cdr1Length = cdr1LengthDic[ID]
133 for c in absentAACDR1Dic[cdr1Length]:
134 absentAAbyID[c] -= 1
135
136 cdr2Length = cdr2LengthDic[ID]
137 for c in absentAACDR2Dic[cdr2Length]:
138 absentAAbyID[c] -= 1
139 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
140
141
142
143 aa_mutations_file = outfile[:outfile.rindex("/")] + "/aa_mutations.txt"
144 with open(aa_mutations_file, 'w') as o:
145 o.write("row.name\t" + "\t".join([str(x) for x in range(1, AALength-1)]) + "\n")
146 o.write("mutations.at.position\t" + "\t".join([str(x) for x in AA_mutation[1:]]) + "\n")
147 o.write("AA.at.position\t" + "\t".join([str(x) for x in absentAA]) + "\n")
148
149 if linecount == 0:
150 print "No data, exiting"
151 with open(outfile, 'w') as o:
152 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
153 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
154 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
155 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
156 import sys
157
158 sys.exit()
159
160 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
161 RGYWCount = {g: 0 for g in genes}
162 WRCYCount = {g: 0 for g in genes}
163 WACount = {g: 0 for g in genes}
164 TWCount = {g: 0 for g in genes}
165
166 IDIndex = 0
167 ataIndex = 0
168 tatIndex = 0
169 aggctatIndex = 0
170 atagcctIndex = 0
171 first = True
172 with open(infile, 'r') as i:
173 for line in i:
174 if first:
175 linesplt = line.split("\t")
176 ataIndex = linesplt.index("X.a.t.a")
177 tatIndex = linesplt.index("t.a.t.")
178 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
179 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
180 first = False
181 continue
182 linesplt = line.split("\t")
183 gene = linesplt[best_matchIndex]
184 ID = linesplt[IDIndex]
185 RGYW = [(int(x), int(y), z) for (x, y, z) in
186 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
187 WRCY = [(int(x), int(y), z) for (x, y, z) in
188 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
189 WA = [(int(x), int(y), z) for (x, y, z) in
190 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
191 TW = [(int(x), int(y), z) for (x, y, z) in
192 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
193 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
194
195 mutationList = (mutationdic[ID + "_FR1"] if include_fr1 else []) + mutationdic[ID + "_CDR1"] + mutationdic[
196 ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
197 for mutation in mutationList:
198 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
199 mutation_in_RGYW = any([(start <= int(where) <= end) for (start, end, region) in RGYW])
200 mutation_in_WRCY = any([(start <= int(where) <= end) for (start, end, region) in WRCY])
201 mutation_in_WA = any([(start <= int(where) <= end) for (start, end, region) in WA])
202 mutation_in_TW = any([(start <= int(where) <= end) for (start, end, region) in TW])
203
204 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
205
206 if in_how_many_motifs > 0:
207 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
208 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
209 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
210 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
211
212
213 def mean(lst):
214 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
215
216
217 def median(lst):
218 lst = sorted(lst)
219 l = len(lst)
220 if l == 0:
221 return 0
222 if l == 1:
223 return lst[0]
224
225 l = int(l / 2)
226
227 if len(lst) % 2 == 0:
228 print "list length is", l
229 return float(lst[l] + lst[(l - 1)]) / 2.0
230 else:
231 return lst[l]
232
233 funcs = {"mean": mean, "median": median, "sum": sum}
234
235 directory = outfile[:outfile.rfind("/") + 1]
236 value = 0
237 valuedic = dict()
238
239 for fname in funcs.keys():
240 for gene in genes:
241 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v:
242 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip())
243 with open(directory + "all_" + fname + "_value.txt", 'r') as v:
244 valuedic["total_" + fname] = float(v.readlines()[0].rstrip())
245
246 print valuedic
247
248 def get_xyz(lst, gene, f, fname):
249 x = int(round(f(lst)))
250 y = valuedic[gene + "_" + fname]
251 z = str(round(x / float(valuedic[gene + "_" + fname]) * 100, 1)) if valuedic[gene + "_" + fname] != 0 else "0"
252 return (str(x), str(y), z)
253
254 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
255 arr = ["RGYW", "WRCY", "WA", "TW"]
256
257 for fname in funcs.keys():
258 func = funcs[fname]
259 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt"
260 with open(foutfile, 'w') as o:
261 for typ in arr:
262 o.write(typ + " (%)")
263 curr = dic[typ]
264 for gene in genes:
265 geneMatcher = re.compile(".*" + gene + ".*")
266 if valuedic[gene + "_" + fname] is 0:
267 o.write(",0,0,0")
268 else:
269 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
270 o.write("," + x + "," + y + "," + z)
271 # for total
272 x, y, z = get_xyz([y for x, y in curr.iteritems()], "total", func, fname)
273 o.write("," + x + "," + y + "," + z + "\n")
274
275
276 # for testing
277 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
278 with open(seq_motif_file, 'w') as o:
279 o.write("ID\tRGYWC\tWRCY\tWA\tTW\n")
280 for ID in IDlist:
281 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")