0
|
1
|
|
2 #
|
|
3 # $Id: Swissprot.pm,v 1.11 2002/10/22 07:38:33 lapp Exp $
|
|
4 #
|
|
5 # BioPerl module for Bio::Index::Abstract
|
|
6 #
|
|
7 # Cared for by Ewan Birney <birney@sanger.ac.uk>
|
|
8 #
|
|
9 # You may distribute this module under the same terms as perl itself
|
|
10
|
|
11 # POD documentation - main docs before the code
|
|
12
|
|
13 =head1 NAME
|
|
14
|
|
15 Bio::Index::Swissprot - Interface for indexing (multiple) Swissprot
|
|
16 .dat files (ie flat file swissprot format).
|
|
17
|
|
18 =head1 SYNOPSIS
|
|
19
|
|
20 # Complete code for making an index for several
|
|
21 # Swissprot files
|
|
22 use Bio::Index::Swissprot;
|
|
23 use strict;
|
|
24
|
|
25 my $Index_File_Name = shift;
|
|
26 my $inx = Bio::Index::Swissprot->new('-filename' => $Index_File_Name,
|
|
27 '-write_flag' => 'WRITE');
|
|
28 $inx->make_index(@ARGV);
|
|
29
|
|
30 # Print out several sequences present in the index
|
|
31 # in gcg format
|
|
32 use Bio::Index::Swissprot;
|
|
33 use Bio::SeqIO;
|
|
34 use strict;
|
|
35
|
|
36 my $out = Bio::SeqIO->new( '-format' => 'gcg', '-fh' => \*STDOUT );
|
|
37 my $Index_File_Name = shift;
|
|
38 my $inx = Bio::Index::Swissprot->new('-filename' => $Index_File_Name);
|
|
39
|
|
40 foreach my $id (@ARGV) {
|
|
41 my $seq = $inx->fetch($id); # Returns Bio::Seq object
|
|
42 $out->write_seq($seq);
|
|
43 }
|
|
44
|
|
45 # alternatively
|
|
46
|
|
47 my $seq1 = $inx->get_Seq_by_id($id);
|
|
48 my $seq2 = $inx->get_Seq_by_acc($acc);
|
|
49
|
|
50 =head1 DESCRIPTION
|
|
51
|
|
52 Inherits functions for managing dbm files from Bio::Index::Abstract.pm,
|
|
53 and provides the basic funtionallity for indexing Swissprot files, and
|
|
54 retrieving the sequence from them. Heavily snaffled from James Gilbert's
|
|
55 Fasta system. Note: for best results 'use strict'.
|
|
56
|
|
57 =head1 FEED_BACK
|
|
58
|
|
59 =head2 Mailing Lists
|
|
60
|
|
61 User feedback is an integral part of the evolution of this and other
|
|
62 Bioperl modules. Send your comments and suggestions preferably to one
|
|
63 of the Bioperl mailing lists. Your participation is much appreciated.
|
|
64
|
|
65 bioperl-l@bioperl.org - General discussion
|
|
66 http://bio.perl.org/MailList.html - About the mailing lists
|
|
67
|
|
68 =head2 Reporting Bugs
|
|
69
|
|
70 Report bugs to the Bioperl bug tracking system to help us keep track
|
|
71 the bugs and their resolution. Bug reports can be submitted via
|
|
72 email or the web:
|
|
73
|
|
74 bioperl-bugs@bio.perl.org
|
|
75 http://bugzilla.bioperl.org/
|
|
76
|
|
77 =head1 AUTHOR - Ewan Birney
|
|
78
|
|
79 Email - birney@sanger.ac.uk
|
|
80 (Swissprot adaption: lorenz@ist.org)
|
|
81
|
|
82 =head1 APPENDIX
|
|
83
|
|
84 The rest of the documentation details each of the object methods.
|
|
85 Internal methods are usually preceded with a _
|
|
86
|
|
87 =cut
|
|
88
|
|
89
|
|
90 # Let's begin the code...
|
|
91
|
|
92
|
|
93 package Bio::Index::Swissprot;
|
|
94
|
|
95 use vars qw($VERSION @ISA);
|
|
96 use strict;
|
|
97
|
|
98 use Bio::Index::AbstractSeq;
|
|
99 use Bio::Seq;
|
|
100
|
|
101 @ISA = qw(Bio::Index::AbstractSeq);
|
|
102
|
|
103 sub _type_stamp {
|
|
104 return '__Swissprot_FLAT__'; # What kind of index are we?
|
|
105 }
|
|
106
|
|
107 #
|
|
108 # Suggested fix by Michael G Schwern <schwern@pobox.com> to
|
|
109 # get around a clash with CPAN shell...
|
|
110 #
|
|
111
|
|
112 BEGIN {
|
|
113 $VERSION = 0.1;
|
|
114 }
|
|
115
|
|
116 sub _version {
|
|
117 return $VERSION;
|
|
118 }
|
|
119
|
|
120 =head2 _index_file
|
|
121
|
|
122 Title : _index_file
|
|
123 Usage : $index->_index_file( $file_name, $i )
|
|
124 Function: Specialist function to index Swissprot format files.
|
|
125 Is provided with a filename and an integer
|
|
126 by make_index in its SUPER class.
|
|
127 Example :
|
|
128 Returns :
|
|
129 Args :
|
|
130
|
|
131 =cut
|
|
132
|
|
133 sub _index_file {
|
|
134 my( $self,
|
|
135 $file, # File name
|
|
136 $i # Index-number of file being indexed
|
|
137 ) = @_;
|
|
138
|
|
139 my( $begin, # Offset from start of file of the start
|
|
140 # of the last found record.
|
|
141 $id, # ID of last found record.
|
|
142 @accs, # accession of last record. Also put into the index
|
|
143 );
|
|
144
|
|
145 $begin = 0;
|
|
146
|
|
147 open SWISSPROT, $file or $self->throw("Can't open file for read : $file");
|
|
148
|
|
149 # Main indexing loop
|
|
150 $id = undef;
|
|
151 @accs = ();
|
|
152 while (<SWISSPROT>) {
|
|
153 if( /^\/\// ) {
|
|
154 if( ! defined $id ) {
|
|
155 $self->throw("Got to a end of entry line for an Swissprot flat file with no parsed ID. Considering this a problem!");
|
|
156 next;
|
|
157 }
|
|
158 if( ! @accs ) {
|
|
159 $self->warn("For id [$id] in Swissprot flat file, got no accession number. Storing id index anyway");
|
|
160 }
|
|
161
|
|
162 $self->add_record($id, $i, $begin);
|
|
163
|
|
164 foreach my $acc (@accs) {
|
|
165 if( $acc ne $id ) {
|
|
166 $self->add_record($acc, $i, $begin);
|
|
167 }
|
|
168 }
|
|
169 @accs = (); # reset acc array
|
|
170 $id = undef; # reset id
|
|
171 } elsif (/^ID\s+(\S+)/) {
|
|
172 $id = $1;
|
|
173 # not sure if I like this. Assummes tell is in bytes.
|
|
174 # we could tell before each line and save it.
|
|
175 $begin = tell(SWISSPROT) - length( $_ );
|
|
176
|
|
177 } elsif (/^AC(.*)/) { # ignore ? if there.
|
|
178 push(@accs, ($1 =~ /\s*(\S+);/g));
|
|
179 } else {
|
|
180 # do nothing
|
|
181 }
|
|
182 }
|
|
183
|
|
184 close SWISSPROT;
|
|
185 return 1;
|
|
186 }
|
|
187
|
|
188 =head2 _file_format
|
|
189
|
|
190 Title : _file_format
|
|
191 Usage : Internal function for indexing system
|
|
192 Function: Provides file format for this database
|
|
193 Example :
|
|
194 Returns :
|
|
195 Args :
|
|
196
|
|
197
|
|
198 =cut
|
|
199
|
|
200 sub _file_format{
|
|
201 my ($self,@args) = @_;
|
|
202
|
|
203 return 'swiss';
|
|
204 }
|
|
205
|
|
206
|
|
207
|
|
208 1;
|
|
209
|
|
210
|
|
211
|
|
212
|
|
213
|
|
214
|
|
215
|
|
216
|
|
217
|
|
218
|