|
0
|
1 # $Id: SequenceProcessorI.pm,v 1.1 2002/10/24 18:35:46 lapp Exp $
|
|
|
2 #
|
|
|
3 # BioPerl module for Bio::Factory::SequenceProcessorI
|
|
|
4 #
|
|
|
5 # Cared for by Hilmar Lapp <hlapp at gmx.net>
|
|
|
6 #
|
|
|
7 # Copyright Hilmar Lapp
|
|
|
8 #
|
|
|
9 # You may distribute this module under the same terms as perl itself
|
|
|
10
|
|
|
11 #
|
|
|
12 # (c) Hilmar Lapp, hlapp at gmx.net, 2002.
|
|
|
13 # (c) GNF, Genomics Institute of the Novartis Research Foundation, 2002.
|
|
|
14 #
|
|
|
15 # You may distribute this module under the same terms as perl itself.
|
|
|
16 # Refer to the Perl Artistic License (see the license accompanying this
|
|
|
17 # software package, or see http://www.perl.com/language/misc/Artistic.html)
|
|
|
18 # for the terms under which you may use, modify, and redistribute this module.
|
|
|
19 #
|
|
|
20 # THIS PACKAGE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED
|
|
|
21 # WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
|
|
|
22 # MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
|
|
|
23 #
|
|
|
24
|
|
|
25 # POD documentation - main docs before the code
|
|
|
26
|
|
|
27 =head1 NAME
|
|
|
28
|
|
|
29 Bio::Factory::SequenceProcessorI - Interface for chained sequence
|
|
|
30 processing algorithms
|
|
|
31
|
|
|
32 =head1 SYNOPSIS
|
|
|
33
|
|
|
34 use Bio::SeqIO;
|
|
|
35 use MySeqProcessor; # is-a Bio::Factory::SequenceProcessorI
|
|
|
36
|
|
|
37 # obtain your source stream, e.g., an EMBL file
|
|
|
38 my $seqin = Bio::SeqIO->new(-fh => \*STDIN, -format => 'embl');
|
|
|
39 # create your processor (it must implement this interface)
|
|
|
40 my $seqalgo = MySeqProcessor->new();
|
|
|
41 # chain together
|
|
|
42 $seqalgo->source_stream($seqin);
|
|
|
43 # you could create more processors and chain them one after another
|
|
|
44 # ...
|
|
|
45 # finally, the last link in the chain is your SeqIO stream
|
|
|
46 my $seqpipe = $seqalgo;
|
|
|
47
|
|
|
48 # once you've established the pipeline, proceed as if you had a
|
|
|
49 # single SeqIO stream
|
|
|
50 while(my $seq = $seqpipe->next_seq()) {
|
|
|
51 # ... do something ...
|
|
|
52 }
|
|
|
53
|
|
|
54 =head1 DESCRIPTION
|
|
|
55
|
|
|
56 This defines an interface that allows seamless chaining of sequence
|
|
|
57 processing algorithms encapsulated in modules while retaining the
|
|
|
58 overall Bio::SeqIO interface at the end of the pipeline.
|
|
|
59
|
|
|
60 This is especially useful if you want an easily configurable
|
|
|
61 processing pipeline of re-usable algorithms as building blocks instead
|
|
|
62 of (hard-)coding the whole algorithm in a single script.
|
|
|
63
|
|
|
64 There are literally no restrictions as to what an individual module
|
|
|
65 can do with a sequence object it obtains from the source stream before
|
|
|
66 it makes it available through its own next_seq() method. It can
|
|
|
67 manipulate the sequence object, but otherwise keep it intact, but it
|
|
|
68 can also create any number of new sequence objects from it, or it can
|
|
|
69 discard some, or any combination thereof. The only requirement is that
|
|
|
70 its next_seq() method return Bio::PrimarySeqI compliant objects. In
|
|
|
71 order to play nice, if a processor creates new objects it should try
|
|
|
72 to use the same sequence factory that the source stream uses, but this
|
|
|
73 is not strongly mandated.
|
|
|
74
|
|
|
75 =head1 FEEDBACK
|
|
|
76
|
|
|
77 =head2 Mailing Lists
|
|
|
78
|
|
|
79 User feedback is an integral part of the evolution of this and other
|
|
|
80 Bioperl modules. Send your comments and suggestions preferably to
|
|
|
81 the Bioperl mailing list. Your participation is much appreciated.
|
|
|
82
|
|
|
83 bioperl-l@bioperl.org - General discussion
|
|
|
84 http://bioperl.org/MailList.shtml - About the mailing lists
|
|
|
85
|
|
|
86 =head2 Reporting Bugs
|
|
|
87
|
|
|
88 Report bugs to the Bioperl bug tracking system to help us keep track
|
|
|
89 of the bugs and their resolution. Bug reports can be submitted via
|
|
|
90 email or the web:
|
|
|
91
|
|
|
92 bioperl-bugs@bioperl.org
|
|
|
93 http://bioperl.org/bioperl-bugs/
|
|
|
94
|
|
|
95 =head1 AUTHOR - Hilmar Lapp
|
|
|
96
|
|
|
97 Email hlapp at gmx.net
|
|
|
98
|
|
|
99 Describe contact details here
|
|
|
100
|
|
|
101 =head1 CONTRIBUTORS
|
|
|
102
|
|
|
103 Additional contributors names and emails here
|
|
|
104
|
|
|
105 =head1 APPENDIX
|
|
|
106
|
|
|
107 The rest of the documentation details each of the object methods.
|
|
|
108 Internal methods are usually preceded with a _
|
|
|
109
|
|
|
110 =cut
|
|
|
111
|
|
|
112
|
|
|
113 # Let the code begin...
|
|
|
114
|
|
|
115
|
|
|
116 package Bio::Factory::SequenceProcessorI;
|
|
|
117 use vars qw(@ISA);
|
|
|
118 use strict;
|
|
|
119 use Carp;
|
|
|
120 use Bio::Root::RootI;
|
|
|
121 use Bio::Factory::SequenceStreamI;
|
|
|
122
|
|
|
123 @ISA = qw( Bio::Factory::SequenceStreamI );
|
|
|
124
|
|
|
125 =head2 source_stream
|
|
|
126
|
|
|
127 Title : source_stream
|
|
|
128 Usage : $obj->source_stream($newval)
|
|
|
129 Function: Get/set the source sequence stream for this sequence
|
|
|
130 processor.
|
|
|
131
|
|
|
132 An implementation is not required to allow set, but will
|
|
|
133 usually do so.
|
|
|
134
|
|
|
135 Example :
|
|
|
136 Returns : A Bio::Factory::SequenceStreamI compliant object
|
|
|
137 Args : on set, new value (a Bio::Factory::SequenceStreamI compliant
|
|
|
138 object)
|
|
|
139
|
|
|
140
|
|
|
141 =cut
|
|
|
142
|
|
|
143 sub source_stream{
|
|
|
144 shift->throw_not_implemented();
|
|
|
145 }
|
|
|
146
|
|
|
147 =head1 Bio::Factory::SequenceStreamI methods
|
|
|
148
|
|
|
149 The requirement to implement these methods is inherited from
|
|
|
150 L<Bio::Factory::SequenceStreamI>. An implementation may not
|
|
|
151 necessarily have to implement all methods in a meaningful way. Which
|
|
|
152 methods will be necessary very much depends on the context in which
|
|
|
153 an implementation of this interface is used. E.g., if it is only used
|
|
|
154 for post-processing sequences read from a SeqIO stream, write_seq()
|
|
|
155 will not be used and hence does not need to be implemented in a
|
|
|
156 meaningful way (it may in fact even throw an exception).
|
|
|
157
|
|
|
158 Also, since an implementor will already receive built objects from a
|
|
|
159 sequence stream, sequence_factory() may or may not be relevant,
|
|
|
160 depending on whether the processing method does or does not involve
|
|
|
161 creating new objects.
|
|
|
162
|
|
|
163 =cut
|
|
|
164
|
|
|
165 =head2 next_seq
|
|
|
166
|
|
|
167 Title : next_seq
|
|
|
168 Usage : $seq = stream->next_seq
|
|
|
169 Function: Reads the next sequence object from the stream and returns it.
|
|
|
170
|
|
|
171 In the case of a non-recoverable situation an exception
|
|
|
172 will be thrown. Do not assume that you can resume parsing
|
|
|
173 the same stream after catching the exception. Note that you
|
|
|
174 can always turn recoverable errors into exceptions by
|
|
|
175 calling $stream->verbose(2).
|
|
|
176
|
|
|
177 Returns : a Bio::Seq sequence object
|
|
|
178 Args : none
|
|
|
179
|
|
|
180 See L<Bio::Root::RootI>
|
|
|
181
|
|
|
182 =cut
|
|
|
183
|
|
|
184 =head2 write_seq
|
|
|
185
|
|
|
186 Title : write_seq
|
|
|
187 Usage : $stream->write_seq($seq)
|
|
|
188 Function: writes the $seq object into the stream
|
|
|
189 Returns : 1 for success and 0 for error
|
|
|
190 Args : Bio::Seq object
|
|
|
191
|
|
|
192 =cut
|
|
|
193
|
|
|
194 =head2 sequence_factory
|
|
|
195
|
|
|
196 Title : sequence_factory
|
|
|
197 Usage : $seqio->sequence_factory($seqfactory)
|
|
|
198 Function: Get the Bio::Factory::SequenceFactoryI
|
|
|
199 Returns : Bio::Factory::SequenceFactoryI
|
|
|
200 Args : none
|
|
|
201
|
|
|
202
|
|
|
203 =cut
|
|
|
204
|
|
|
205 1;
|
