Mercurial > repos > davidvanzessen > shm_csr

comparison shm_csr.py @ 0:c33d93683a09 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Thu, 13 Oct 2016 10:52:24 -0400
parents
children faae21ba5c63
comparison
equal deleted inserted replaced
-1:000000000000 0:c33d93683a09
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="Should the mutation/nucleotides in the FR1 region be included?")
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 NAMatchResult = (None, None, None, None, None, None, '')
26 linecount = 0
27
28 IDIndex = 0
29 best_matchIndex = 0
30 fr1Index = 0
31 cdr1Index = 0
32 fr2Index = 0
33 cdr2Index = 0
34 fr3Index = 0
35 first = True
36 IDlist = []
37 mutationList = []
38 mutationListByID = {}
39 cdr1LengthDic = {}
40 cdr2LengthDic = {}
41
42 with open(infile, 'r') as i:
43 for line in i:
44 if first:
45 linesplt = line.split("\t")
46 IDIndex = linesplt.index("Sequence.ID")
47 best_matchIndex = linesplt.index("best_match")
48 fr1Index = linesplt.index("FR1.IMGT")
49 cdr1Index = linesplt.index("CDR1.IMGT")
50 fr2Index = linesplt.index("FR2.IMGT")
51 cdr2Index = linesplt.index("CDR2.IMGT")
52 fr3Index = linesplt.index("FR3.IMGT")
53 cdr1LengthIndex = linesplt.index("CDR1.IMGT.length")
54 cdr2LengthIndex = linesplt.index("CDR2.IMGT.length")
55 first = False
56 continue
57 linecount += 1
58 linesplt = line.split("\t")
59 ID = linesplt[IDIndex]
60 genedic[ID] = linesplt[best_matchIndex]
61 try:
62 if linesplt[fr1Index] != "NA":
63 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x] if include_fr1 else []
64 else:
65 mutationdic[ID + "_FR1"] = []
66 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x] if linesplt[cdr1Index] != "NA" else []
67 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x] if linesplt[fr2Index] != "NA" else []
68 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x] if linesplt[cdr2Index] != "NA" else []
69 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
70 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x] if linesplt[fr3Index] != "NA" else []
71 except e:
72 print linesplt
73 print linecount
74 print e
75 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
76 mutationListByID[ID] = mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
77
78 cdr1Length = linesplt[cdr1LengthIndex]
79 cdr2Length = linesplt[cdr2LengthIndex]
80
81 cdr1LengthDic[ID] = int(cdr1Length) if cdr1Length != "X" else 0
82 cdr2LengthDic[ID] = int(cdr2Length) if cdr2Length != "X" else 0
83
84 IDlist += [ID]
85
86 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
87 if AALength < 60:
88 AALength = 64
89
90 AA_mutation = [0] * AALength
91 AA_mutation_dic = {"IGA": AA_mutation[:], "IGG": AA_mutation[:], "IGM": AA_mutation[:], "unm": AA_mutation[:]}
92 AA_mutation_empty = AA_mutation[:]
93
94 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/aa_id_mutations.txt"
95 with open(aa_mutations_by_id_file, 'w') as o:
96 o.write("ID\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
97 for ID in mutationListByID.keys():
98 AA_mutation_for_ID = AA_mutation_empty[:]
99 for mutation in mutationListByID[ID]:
100 if mutation[4]:
101 AA_mutation_position = int(mutation[4])
102 AA_mutation[AA_mutation_position] += 1
103 AA_mutation_for_ID[AA_mutation_position] += 1
104 clss = genedic[ID][:3]
105 AA_mutation_dic[clss][AA_mutation_position] += 1
106 o.write(ID + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in AA_mutation_for_ID[1:]]) + "\n")
107
108
109
110 #absent AA stuff
111 absentAACDR1Dic = defaultdict(list)
112 absentAACDR1Dic[5] = range(29,36)
113 absentAACDR1Dic[6] = range(29,35)
114 absentAACDR1Dic[7] = range(30,35)
115 absentAACDR1Dic[8] = range(30,34)
116 absentAACDR1Dic[9] = range(31,34)
117 absentAACDR1Dic[10] = range(31,33)
118 absentAACDR1Dic[11] = [32]
119
120 absentAACDR2Dic = defaultdict(list)
121 absentAACDR2Dic[0] = range(55,65)
122 absentAACDR2Dic[1] = range(56,65)
123 absentAACDR2Dic[2] = range(56,64)
124 absentAACDR2Dic[3] = range(57,64)
125 absentAACDR2Dic[4] = range(57,63)
126 absentAACDR2Dic[5] = range(58,63)
127 absentAACDR2Dic[6] = range(58,62)
128 absentAACDR2Dic[7] = range(59,62)
129 absentAACDR2Dic[8] = range(59,61)
130 absentAACDR2Dic[9] = [60]
131
132 absentAA = [len(IDlist)] * (AALength-1)
133 for k, cdr1Length in cdr1LengthDic.iteritems():
134 for c in absentAACDR1Dic[cdr1Length]:
135 absentAA[c] -= 1
136
137 for k, cdr2Length in cdr2LengthDic.iteritems():
138 for c in absentAACDR2Dic[cdr2Length]:
139 absentAA[c] -= 1
140
141
142 aa_mutations_by_id_file = outfile[:outfile.rindex("/")] + "/absent_aa_id.txt"
143 with open(aa_mutations_by_id_file, 'w') as o:
144 o.write("ID\tcdr1length\tcdr2length\tbest_match\t" + "\t".join([str(x) for x in range(1,AALength)]) + "\n")
145 for ID in IDlist:
146 absentAAbyID = [1] * (AALength-1)
147 cdr1Length = cdr1LengthDic[ID]
148 for c in absentAACDR1Dic[cdr1Length]:
149 absentAAbyID[c] -= 1
150
151 cdr2Length = cdr2LengthDic[ID]
152 for c in absentAACDR2Dic[cdr2Length]:
153 absentAAbyID[c] -= 1
154 o.write(ID + "\t" + str(cdr1Length) + "\t" + str(cdr2Length) + "\t" + genedic[ID] + "\t" + "\t".join([str(x) for x in absentAAbyID]) + "\n")
155
156 if linecount == 0:
157 print "No data, exiting"
158 with open(outfile, 'w') as o:
159 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
160 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
161 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
162 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
163 import sys
164
165 sys.exit()
166
167 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
168 RGYWCount = {}
169 WRCYCount = {}
170 WACount = {}
171 TWCount = {}
172
173 #IDIndex = 0
174 ataIndex = 0
175 tatIndex = 0
176 aggctatIndex = 0
177 atagcctIndex = 0
178 first = True
179 with open(infile, 'r') as i:
180 for line in i:
181 if first:
182 linesplt = line.split("\t")
183 ataIndex = linesplt.index("X.a.t.a")
184 tatIndex = linesplt.index("t.a.t.")
185 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
186 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
187 first = False
188 continue
189 linesplt = line.split("\t")
190 gene = linesplt[best_matchIndex]
191 ID = linesplt[IDIndex]
192 if ID == "ca2":
193 print linesplt
194 RGYW = [(int(x), int(y), z) for (x, y, z) in
195 [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
196 WRCY = [(int(x), int(y), z) for (x, y, z) in
197 [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
198 WA = [(int(x), int(y), z) for (x, y, z) in
199 [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
200 TW = [(int(x), int(y), z) for (x, y, z) in
201 [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
202 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0, 0, 0, 0
203
204 mutationList = (mutationdic[ID + "_FR1"] if include_fr1 else []) + mutationdic[ID + "_CDR1"] + mutationdic[
205 ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
206 for mutation in mutationList:
207 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
208 mutation_in_RGYW = any([(start <= int(where) <= end) for (start, end, region) in RGYW])
209 mutation_in_WRCY = any([(start <= int(where) <= end) for (start, end, region) in WRCY])
210 mutation_in_WA = any([(start <= int(where) <= end) for (start, end, region) in WA])
211 mutation_in_TW = any([(start <= int(where) <= end) for (start, end, region) in TW])
212
213 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
214
215 if in_how_many_motifs > 0:
216 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
217 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
218 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
219 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
220
221
222 def mean(lst):
223 return (float(sum(lst)) / len(lst)) if len(lst) > 0 else 0.0
224
225
226 def median(lst):
227 lst = sorted(lst)
228 l = len(lst)
229 if l == 0:
230 return 0
231 if l == 1:
232 return lst[0]
233
234 l = int(l / 2)
235
236 if len(lst) % 2 == 0:
237 return float(lst[l] + lst[(l - 1)]) / 2.0
238 else:
239 return lst[l]
240
241 funcs = {"mean": mean, "median": median, "sum": sum}
242
243 directory = outfile[:outfile.rfind("/") + 1]
244 value = 0
245 valuedic = dict()
246
247 for fname in funcs.keys():
248 for gene in genes:
249 with open(directory + gene + "_" + fname + "_value.txt", 'r') as v:
250 valuedic[gene + "_" + fname] = float(v.readlines()[0].rstrip())
251 with open(directory + "all_" + fname + "_value.txt", 'r') as v:
252 valuedic["total_" + fname] = float(v.readlines()[0].rstrip())
253
254
255 def get_xyz(lst, gene, f, fname):
256 x = int(round(f(lst)))
257 y = valuedic[gene + "_" + fname]
258 z = str(round(x / float(y) * 100, 1)) if y != 0 else "0"
259 return (str(x), str(y), z)
260
261 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
262 arr = ["RGYW", "WRCY", "WA", "TW"]
263
264 geneMatchers = {gene: re.compile("^" + gene + ".*") for gene in genes}
265
266 for fname in funcs.keys():
267 func = funcs[fname]
268 foutfile = outfile[:outfile.rindex("/")] + "/hotspot_analysis_" + fname + ".txt"
269 with open(foutfile, 'w') as o:
270 for typ in arr:
271 o.write(typ + " (%)")
272 curr = dic[typ]
273 for gene in genes:
274 geneMatcher = geneMatchers[gene] #re.compile("^" + gene + ".*") #recompile every loop....
275 if valuedic[gene + "_" + fname] is 0:
276 o.write(",0,0,0")
277 else:
278 x, y, z = get_xyz([curr[x] for x in [y for y, z in genedic.iteritems() if geneMatcher.match(z)]], gene, func, fname)
279 o.write("," + x + "," + y + "," + z)
280
281 x, y, z = get_xyz([y for x, y in curr.iteritems() if not genedic[x].startswith("unmatched")], "total", func, fname)
282 o.write("," + x + "," + y + "," + z + "\n")
283
284
285 # for testing
286 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
287 with open(seq_motif_file, 'w') as o:
288 o.write("ID\tRGYWC\tWRCY\tWA\tTW\n")
289 for ID in IDlist:
290 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")