Mercurial > repos > davidvanzessen > mutation_analysis

comparison imgt_loader.py @ 47:099cc1254f74 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Mon, 09 Nov 2015 07:21:09 -0500
parents
children
comparison
equal deleted inserted replaced
46:42e39033514b 47:099cc1254f74
1 import pandas as pd
2 try:
3 pd.options.mode.chained_assignment = None # default='warn'
4 except:
5 pass
6 import re
7 import argparse
8 import os
9
10 def stop_err( msg, ret=1 ):
11 sys.stderr.write( msg )
12 sys.exit( ret )
13
14 #docs.python.org/dev/library/argparse.html
15 parser = argparse.ArgumentParser()
16 parser.add_argument("--summ", help="The 1_Summary file from the imgt output")
17 parser.add_argument("--aa", help="The 5_AA-Sequence file from the imgt output")
18 parser.add_argument("--junction", help="The 6_Junction file from the imgt output")
19 parser.add_argument("--output", help="Output file")
20
21 args = parser.parse_args()
22
23 old_summary_columns = [u'Sequence ID', u'JUNCTION frame', u'V-GENE and allele', u'D-GENE and allele', u'J-GENE and allele', u'CDR1-IMGT length', u'CDR2-IMGT length', u'CDR3-IMGT length', u'Orientation']
24 old_sequence_columns = [u'CDR1-IMGT', u'CDR2-IMGT', u'CDR3-IMGT']
25 old_junction_columns = [u'JUNCTION']
26
27 added_summary_columns = [u'Functionality', u'V-REGION identity %', u'V-REGION identity nt', u'D-REGION reading frame', u'AA JUNCTION', u'Functionality comment', u'Sequence']
28 added_sequence_columns = [u'FR1-IMGT', u'FR2-IMGT', u'FR3-IMGT', u'CDR3-IMGT', u'JUNCTION', u'J-REGION', u'FR4-IMGT']
29 added_junction_columns = [u"P3'V-nt nb", u'N1-REGION-nt nb', u"P5'D-nt nb", u"P3'D-nt nb", u'N2-REGION-nt nb', u"P5'J-nt nb", u"3'V-REGION trimmed-nt nb", u"5'D-REGION trimmed-nt nb", u"3'D-REGION trimmed-nt nb", u"5'J-REGION trimmed-nt nb"]
30
31 outFile = args.output
32
33 #fSummary = pd.read_csv(triplets[0][0], sep="\t", low_memory=False)
34 fSummary = pd.read_csv(args.summ, sep="\t", dtype=object)
35 #fSequence = pd.read_csv(triplets[0][1], sep="\t", low_memory=False)
36 fSequence = pd.read_csv(args.aa, sep="\t", dtype=object)
37 #fJunction = pd.read_csv(triplets[0][2], sep="\t", low_memory=False)
38 fJunction = pd.read_csv(args.junction, sep="\t", dtype=object)
39 tmp = fSummary[["Sequence ID", "JUNCTION frame", "V-GENE and allele", "D-GENE and allele", "J-GENE and allele"]]
40
41 tmp["CDR1 Seq"] = fSequence["CDR1-IMGT"]
42 tmp["CDR1 Length"] = fSummary["CDR1-IMGT length"]
43
44 tmp["CDR2 Seq"] = fSequence["CDR2-IMGT"]
45 tmp["CDR2 Length"] = fSummary["CDR2-IMGT length"]
46
47 tmp["CDR3 Seq"] = fSequence["CDR3-IMGT"]
48 tmp["CDR3 Length"] = fSummary["CDR3-IMGT length"]
49
50 tmp["CDR3 Seq DNA"] = fJunction["JUNCTION"]
51 tmp["CDR3 Length DNA"] = '1'
52 tmp["Strand"] = fSummary["Orientation"]
53 tmp["CDR3 Found How"] = 'a'
54
55 for col in added_summary_columns:
56 tmp[col] = fSummary[col]
57
58 for col in added_sequence_columns:
59 tmp[col] = fSequence[col]
60
61 for col in added_junction_columns:
62 tmp[col] = fJunction[col]
63
64 outFrame = tmp
65
66 outFrame.columns = [u'ID', u'VDJ Frame', u'Top V Gene', u'Top D Gene', u'Top J Gene', u'CDR1 Seq', u'CDR1 Length', u'CDR2 Seq', u'CDR2 Length', u'CDR3 Seq', u'CDR3 Length', u'CDR3 Seq DNA', u'CDR3 Length DNA', u'Strand', u'CDR3 Found How', u'Functionality', 'V-REGION identity %', 'V-REGION identity nt', 'D-REGION reading frame', 'AA JUNCTION', 'Functionality comment', 'Sequence', 'FR1-IMGT', 'FR2-IMGT', 'FR3-IMGT', 'CDR3-IMGT', 'JUNCTION', 'J-REGION', 'FR4-IMGT', 'P3V-nt nb', 'N1-REGION-nt nb', 'P5D-nt nb', 'P3D-nt nb', 'N2-REGION-nt nb', 'P5J-nt nb', '3V-REGION trimmed-nt nb', '5D-REGION trimmed-nt nb', '3D-REGION trimmed-nt nb', '5J-REGION trimmed-nt nb']
67
68 """
69 IGHV[0-9]-[0-9ab]+-?[0-9]?D?
70 TRBV[0-9]{1,2}-?[0-9]?-?[123]?
71 IGKV[0-3]D?-[0-9]{1,2}
72 IGLV[0-9]-[0-9]{1,2}
73 TRAV[0-9]{1,2}(-[1-46])?(/DV[45678])?
74 TRGV[234589]
75 TRDV[1-3]
76
77 IGHD[0-9]-[0-9ab]+
78 TRBD[12]
79 TRDD[1-3]
80
81 IGHJ[1-6]
82 TRBJ[12]-[1-7]
83 IGKJ[1-5]
84 IGLJ[12367]
85 TRAJ[0-9]{1,2}
86 TRGJP?[12]
87 TRDJ[1-4]
88 """
89
90 vPattern = [r"(IGHV[0-9]-[0-9ab]+-?[0-9]?D?)",
91 r"(TRBV[0-9]{1,2}-?[0-9]?-?[123]?)",
92 r"(IGKV[0-3]D?-[0-9]{1,2})",
93 r"(IGLV[0-9]-[0-9]{1,2})",
94 r"(TRAV[0-9]{1,2}(-[1-46])?(/DV[45678])?)",
95 r"(TRGV[234589])",
96 r"(TRDV[1-3])"]
97
98 dPattern = [r"(IGHD[0-9]-[0-9ab]+)",
99 r"(TRBD[12])",
100 r"(TRDD[1-3])"]
101
102 jPattern = [r"(IGHJ[1-6])",
103 r"(TRBJ[12]-[1-7])",
104 r"(IGKJ[1-5])",
105 r"(IGLJ[12367])",
106 r"(TRAJ[0-9]{1,2})",
107 r"(TRGJP?[12])",
108 r"(TRDJ[1-4])"]
109
110 vPattern = re.compile(r"|".join(vPattern))
111
112 dPattern = re.compile(r"|".join(dPattern))
113
114 jPattern = re.compile(r"|".join(jPattern))
115
116
117 def filterGenes(s, pattern):
118 if type(s) is not str:
119 return "NA"
120 res = pattern.search(s)
121 if res:
122 return res.group(0)
123 return "NA"
124
125
126
127 outFrame["Top V Gene"] = outFrame["Top V Gene"].apply(lambda x: filterGenes(x, vPattern))
128 outFrame["Top D Gene"] = outFrame["Top D Gene"].apply(lambda x: filterGenes(x, dPattern))
129 outFrame["Top J Gene"] = outFrame["Top J Gene"].apply(lambda x: filterGenes(x, jPattern))
130
131
132 tmp = outFrame["VDJ Frame"]
133 tmp = tmp.replace("in-frame", "In-frame")
134 tmp = tmp.replace("null", "Out-of-frame")
135 tmp = tmp.replace("out-of-frame", "Out-of-frame")
136 outFrame["VDJ Frame"] = tmp
137 outFrame["CDR3 Length DNA"] = outFrame["CDR3 Seq DNA"].map(str).map(len)
138 safeLength = lambda x: len(x) if type(x) == str else 0
139 #outFrame = outFrame[(outFrame["CDR3 Seq DNA"].map(safeLength) > 0) & (outFrame["Top V Gene"] != "NA") & (outFrame["Top J Gene"] != "NA")] #filter out weird rows?
140 #outFrame = outFrame[(outFrame["CDR3 Seq DNA"].map(safeLength) > 0) & (outFrame["Top V Gene"] != "NA") & (outFrame["Top D Gene"] != "NA") & (outFrame["Top J Gene"] != "NA")] #filter out weird rows?
141 outFrame.to_csv(outFile, sep="\t", index=False, index_label="index")