comparison imgt_loader/imgt_loader.py @ 0:afe85eb6572e draft

Uploaded
author davidvanzessen
date Mon, 29 Aug 2016 05:41:20 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:afe85eb6572e
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'N-REGION-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",
30 u"5'D-REGION trimmed-nt nb", u"3'D-REGION trimmed-nt nb", u"5'J-REGION trimmed-nt nb", u"N-REGION", u"N1-REGION", u"N2-REGION"]
31
32 outFile = args.output
33
34 #fSummary = pd.read_csv(triplets[0][0], sep="\t", low_memory=False)
35 fSummary = pd.read_csv(args.summ, sep="\t", dtype=object)
36 #fSequence = pd.read_csv(triplets[0][1], sep="\t", low_memory=False)
37 fSequence = pd.read_csv(args.aa, sep="\t", dtype=object)
38 #fJunction = pd.read_csv(triplets[0][2], sep="\t", low_memory=False)
39 fJunction = pd.read_csv(args.junction, sep="\t", dtype=object)
40 tmp = fSummary[["Sequence ID", "JUNCTION frame", "V-GENE and allele", "D-GENE and allele", "J-GENE and allele"]]
41
42 tmp["CDR1 Seq"] = fSequence["CDR1-IMGT"]
43 tmp["CDR1 Length"] = fSummary["CDR1-IMGT length"]
44
45 tmp["CDR2 Seq"] = fSequence["CDR2-IMGT"]
46 tmp["CDR2 Length"] = fSummary["CDR2-IMGT length"]
47
48 tmp["CDR3 Seq"] = fSequence["CDR3-IMGT"]
49 tmp["CDR3 Length"] = fSummary["CDR3-IMGT length"]
50
51 tmp["CDR3 Seq DNA"] = fJunction["JUNCTION"]
52 tmp["CDR3 Length DNA"] = '1'
53 tmp["Strand"] = fSummary["Orientation"]
54 tmp["CDR3 Found How"] = 'a'
55
56 for col in added_summary_columns:
57 tmp[col] = fSummary[col]
58
59 for col in added_sequence_columns:
60 tmp[col] = fSequence[col]
61
62 for col in added_junction_columns:
63 tmp[col] = fJunction[col]
64
65 outFrame = tmp
66
67 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',
68 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',
69 'AA JUNCTION', 'Functionality comment', 'Sequence', 'FR1-IMGT', 'FR2-IMGT', 'FR3-IMGT', 'CDR3-IMGT', 'JUNCTION', 'J-REGION', 'FR4-IMGT', 'P3V-nt nb',
70 'N-REGION-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',
71 '5J-REGION trimmed-nt nb', "N-REGION", "N1-REGION", "N2-REGION"]
72
73 """
74 IGHV[0-9]-[0-9ab]+-?[0-9]?D?
75 TRBV[0-9]{1,2}-?[0-9]?-?[123]?
76 IGKV[0-3]D?-[0-9]{1,2}
77 IGLV[0-9]-[0-9]{1,2}
78 TRAV[0-9]{1,2}(-[1-46])?(/DV[45678])?
79 TRGV[234589]
80 TRDV[1-3]
81
82 IGHD[0-9]-[0-9ab]+
83 TRBD[12]
84 TRDD[1-3]
85
86 IGHJ[1-6]
87 TRBJ[12]-[1-7]
88 IGKJ[1-5]
89 IGLJ[12367]
90 TRAJ[0-9]{1,2}
91 TRGJP?[12]
92 TRDJ[1-4]
93 """
94
95 vPattern = [r"(IGHV[0-9]-[0-9ab]+-?[0-9]?D?)",
96 r"(TRBV[0-9]{1,2}-?[0-9]?-?[123]?)",
97 r"(IGKV[0-3]D?-[0-9]{1,2})",
98 r"(IGLV[0-9]-[0-9]{1,2})",
99 r"(TRAV[0-9]{1,2}(-[1-46])?(/DV[45678])?)",
100 r"(TRGV[234589])",
101 r"(TRDV[1-3])",
102 r"(IGHV[0-9]S[0-9]+)"]
103
104 dPattern = [r"(IGHD[0-9]-[0-9ab]+)",
105 r"(TRBD[12])",
106 r"(TRDD[1-3])"]
107
108 jPattern = [r"(IGHJ[1-6])",
109 r"(TRBJ[12]-[1-7])",
110 r"(IGKJ[1-5])",
111 r"(IGLJ[12367])",
112 r"(TRAJ[0-9]{1,2})",
113 r"(TRGJP?[12])",
114 r"(TRDJ[1-4])"]
115
116 vPattern = re.compile(r"|".join(vPattern))
117
118 dPattern = re.compile(r"|".join(dPattern))
119
120 jPattern = re.compile(r"|".join(jPattern))
121
122
123 def filterGenes(s, pattern):
124 if type(s) is not str:
125 return "NA"
126 res = pattern.search(s)
127 if res:
128 return res.group(0)
129 return "NA"
130
131
132
133 outFrame["Top V Gene"] = outFrame["Top V Gene"].apply(lambda x: filterGenes(x, vPattern))
134 outFrame["Top D Gene"] = outFrame["Top D Gene"].apply(lambda x: filterGenes(x, dPattern))
135 outFrame["Top J Gene"] = outFrame["Top J Gene"].apply(lambda x: filterGenes(x, jPattern))
136
137
138 tmp = outFrame["VDJ Frame"]
139 tmp = tmp.replace("in-frame", "In-frame")
140 tmp = tmp.replace("null", "Out-of-frame")
141 tmp = tmp.replace("out-of-frame", "Out-of-frame")
142 outFrame["VDJ Frame"] = tmp
143 outFrame["CDR3 Length DNA"] = outFrame["CDR3 Seq DNA"].map(str).map(len)
144 safeLength = lambda x: len(x) if type(x) == str else 0
145 #outFrame = outFrame[(outFrame["CDR3 Seq DNA"].map(safeLength) > 0) & (outFrame["Top V Gene"] != "NA") & (outFrame["Top J Gene"] != "NA")] #filter out weird rows?
146 #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?
147 outFrame.to_csv(outFile, sep="\t", index=False, index_label="index")