Mercurial > repos > davidvanzessen > argalaxy_tools

comparison mutation_analysis.py @ 11:0510cf1f7cbc draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Tue, 04 Aug 2015 09:59:26 -0400
parents
children
comparison
equal deleted inserted replaced
10:edbf4fba5fc7 11:0510cf1f7cbc
1 import re
2 import argparse
3
4 parser = argparse.ArgumentParser()
5 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")
6 parser.add_argument("--genes", help="The genes available in the 'best_match' column")
7 parser.add_argument("--includefr1", help="The genes available in the 'best_match' column")
8 parser.add_argument("--output", help="Output file")
9
10 args = parser.parse_args()
11
12 infile = args.input
13 genes = str(args.genes).split(",")
14 print "includefr1 =", args.includefr1
15 include_fr1 = True if args.includefr1 == "yes" else False
16 outfile = args.output
17
18 genedic = dict()
19
20 mutationdic = dict()
21 mutationMatcher = re.compile("^(.)(\d+).(.),?(.)?(\d+)?.?(.)?(.?.?.?.?.?)?")
22 linecount = 0
23
24 IDIndex = 0
25 best_matchIndex = 0
26 fr1Index = 0
27 cdr1Index = 0
28 fr2Index = 0
29 cdr2Index = 0
30 fr3Index = 0
31 first=True
32 IDlist = []
33 mutationList = []
34
35 with open(infile, 'r') as i:
36 for line in i:
37 if first:
38 linesplt = line.split("\t")
39 IDIndex = linesplt.index("Sequence.ID")
40 best_matchIndex = linesplt.index("best_match")
41 fr1Index = linesplt.index("FR1.IMGT")
42 cdr1Index = linesplt.index("CDR1.IMGT")
43 fr2Index = linesplt.index("FR2.IMGT")
44 cdr2Index = linesplt.index("CDR2.IMGT")
45 fr3Index = linesplt.index("FR3.IMGT")
46 first = False
47 continue
48 linecount += 1
49 linesplt = line.split("\t")
50 ID = linesplt[IDIndex]
51 genedic[ID] = linesplt[best_matchIndex]
52 mutationdic[ID + "_FR1"] = [mutationMatcher.match(x).groups() for x in linesplt[fr1Index].split("|") if x] if include_fr1 else []
53 mutationdic[ID + "_CDR1"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr1Index].split("|") if x]
54 mutationdic[ID + "_FR2"] = [mutationMatcher.match(x).groups() for x in linesplt[fr2Index].split("|") if x]
55 mutationdic[ID + "_CDR2"] = [mutationMatcher.match(x).groups() for x in linesplt[cdr2Index].split("|") if x]
56 mutationdic[ID + "_FR2-CDR2"] = mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"]
57 mutationdic[ID + "_FR3"] = [mutationMatcher.match(x).groups() for x in linesplt[fr3Index].split("|") if x]
58
59 mutationList += mutationdic[ID + "_FR1"] + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
60
61 IDlist += [ID]
62
63
64 AA_mutation = [0] * (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
65
66 for mutation in mutationList:
67 if mutation[4]: #if non silent mutation
68 AA_mutation[int(mutation[4])] += 1
69
70 aa_mutations_file = outfile[:outfile.rindex("/")] + "/aa_mutations.txt"
71 with open(aa_mutations_file, 'w') as o:
72 o.write(",".join([str(x) for x in AA_mutation]) + "\n")
73
74 if linecount == 0:
75 print "No data, exiting"
76 with open(outfile, 'w') as o:
77 o.write("RGYW (%)," + ("0,0,0\n" * len(genes)))
78 o.write("WRCY (%)," + ("0,0,0\n" * len(genes)))
79 o.write("WA (%)," + ("0,0,0\n" * len(genes)))
80 o.write("TW (%)," + ("0,0,0\n" * len(genes)))
81 import sys
82 sys.exit()
83
84 hotspotMatcher = re.compile("[actg]+,(\d+)-(\d+)\((.*)\)")
85 RGYWCount = {g: 0 for g in genes}
86 WRCYCount = {g: 0 for g in genes}
87 WACount = {g: 0 for g in genes}
88 TWCount = {g: 0 for g in genes}
89
90 IDIndex = 0
91 ataIndex = 0
92 tatIndex = 0
93 aggctatIndex = 0
94 atagcctIndex = 0
95 first = True
96 with open(infile, 'r') as i:
97 for line in i:
98 if first:
99 linesplt = line.split("\t")
100 ataIndex = linesplt.index("X.a.t.a")
101 tatIndex = linesplt.index("t.a.t.")
102 aggctatIndex = linesplt.index("X.a.g.g.c.t..a.t.")
103 atagcctIndex = linesplt.index("X.a.t..a.g.c.c.t.")
104 first = False
105 continue
106 linesplt = line.split("\t")
107 gene = linesplt[best_matchIndex]
108 ID = linesplt[IDIndex]
109 RGYW = [(int(x),int(y),z) for (x,y,z) in [hotspotMatcher.match(x).groups() for x in linesplt[aggctatIndex].split("|") if x]]
110 WRCY = [(int(x),int(y),z) for (x,y,z) in [hotspotMatcher.match(x).groups() for x in linesplt[atagcctIndex].split("|") if x]]
111 WA = [(int(x),int(y),z) for (x,y,z) in [hotspotMatcher.match(x).groups() for x in linesplt[ataIndex].split("|") if x]]
112 TW = [(int(x),int(y),z) for (x,y,z) in [hotspotMatcher.match(x).groups() for x in linesplt[tatIndex].split("|") if x]]
113 RGYWCount[ID], WRCYCount[ID], WACount[ID], TWCount[ID] = 0,0,0,0
114
115 mutationList = (mutationdic[ID + "_FR1"] if include_fr1 else []) + mutationdic[ID + "_CDR1"] + mutationdic[ID + "_FR2"] + mutationdic[ID + "_CDR2"] + mutationdic[ID + "_FR3"]
116 for mutation in mutationList:
117 frm, where, to, AAfrm, AAwhere, AAto, junk = mutation
118 mutation_in_RGYW = any([(start <= int(where) <= end) for (start,end,region) in RGYW])
119 mutation_in_WRCY = any([(start <= int(where) <= end) for (start,end,region) in WRCY])
120 mutation_in_WA = any([(start <= int(where) <= end) for (start,end,region) in WA])
121 mutation_in_TW = any([(start <= int(where) <= end) for (start,end,region) in TW])
122
123 in_how_many_motifs = sum([mutation_in_RGYW, mutation_in_WRCY, mutation_in_WA, mutation_in_TW])
124
125 if in_how_many_motifs > 0:
126 RGYWCount[ID] += (1.0 * int(mutation_in_RGYW)) / in_how_many_motifs
127 WRCYCount[ID] += (1.0 * int(mutation_in_WRCY)) / in_how_many_motifs
128 WACount[ID] += (1.0 * int(mutation_in_WA)) / in_how_many_motifs
129 TWCount[ID] += (1.0 * int(mutation_in_TW)) / in_how_many_motifs
130
131 directory = outfile[:outfile.rfind("/") + 1]
132 value = 0
133 valuedic = dict()
134 for gene in genes:
135 with open(directory + gene + "_value.txt", 'r') as v:
136 valuedic[gene] = int(v.readlines()[0].rstrip())
137 with open(directory + "total_value.txt", 'r') as v:
138 valuedic["total"] = int(v.readlines()[0].rstrip())
139
140 dic = {"RGYW": RGYWCount, "WRCY": WRCYCount, "WA": WACount, "TW": TWCount}
141 arr = ["RGYW", "WRCY", "WA", "TW"]
142 with open(outfile, 'w') as o:
143 for typ in arr:
144 o.write(typ + " (%)")
145 curr = dic[typ]
146 for gene in genes:
147 geneMatcher = re.compile(".*" + gene + ".*")
148 if valuedic[gene] is 0:
149 o.write(",0,0,0")
150 else:
151 x = int(round(sum([curr[x] for x in [y for y,z in genedic.iteritems() if geneMatcher.match(z)]])))
152 y = valuedic[gene]
153 z = str(round(x / float(valuedic[gene]) * 100, 1))
154 o.write("," + str(x) + "," + str(y) + "," + z)
155 #for total
156 x = int(round(sum([y for x,y in curr.iteritems()])))
157 y = valuedic["total"]
158 z = str(round(x / float(valuedic["total"]) * 100, 1))
159 o.write("," + str(x) + "," + str(y) + "," + z + "\n")
160
161
162 #for testing
163 seq_motif_file = outfile[:outfile.rindex("/")] + "/motif_per_seq.txt"
164 first = True
165 with open(seq_motif_file, 'w') as o:
166 for ID in IDlist:
167 if first:
168 o.write("ID\tRGYWC\tWRCY\tWA\tTW\n")
169 first = False
170 continue
171 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")