bin/bp_index

   1 #!/usr/bin/perl
   2
   3 =head1 NAME
   4
   5 bp_index.pl - indexes files for use by bp_fetch.pl
   6
   7 =head1 SYNOPSIS
   8
   9 bp_index.pl index_name file1 file2 etc.
  10
  11 =head1 DESCRIPTION
  12
  13 bp_index.pl builds a bioperl index for the sequence files given in the
  14 argument list, under the index name. For example
  15
  16    bp_index.pl nrdb /data/nrdb/nrdb.fasta
  17
  18 would build an index called 'nrdb' as the index name for the file
  19 nrdb.fasta, and
  20
  21    bp_index.pl -fmt EMBL swiss /data/swiss/*.dat
  22
  23 would build an index called swiss for all the files in /data/swiss
  24 which end in .dat which are in EMBL format.
  25
  26 The indexes are built using the Bio/Index/* modules, in particular,
  27 Bio::Index::EMBL and the Bio::Index::Fasta modules. Any script which
  28 uses these modules can use the index. A good example script is bp_fetch
  29 which fetches sequences and pipes them to STDOUT, for example
  30
  31    bp_fetch swiss:ROA1_HUMAN
  32
  33 gets the ROA1_HUMAN sequence from the swiss index and writes it as
  34 fasta format on STDOUT.
  35
  36 =head1 OPTIONS
  37
  38   -fmt  <format>   - Fasta (default), swiss or EMBL
  39   -dir  <dir>      - directory where the index files are found
  40                      (overrides BIOPERL_INDEX environment variable)
  41
  42 Options for expert use
  43
  44   -type <db_type>  - DBM_file type.
  45                      (overrides BIOPERL_INDEX_TYPE environment variable)
  46   -v               - report every index addition (debugging)
  47
  48 =head1 ENVIRONMENT
  49
  50 bp_index and bp_fetch coordinate where the databases lie using the
  51 environment variable BIOPERL_INDEX. This can be overridden using the
  52 -dir option. There is no default value, so you must use the -dir option
  53 or set BIOPERL_INDEX.
  54
  55 The DB type is coordinated with BIOPERL_INDEX_TYPE which if it
  56 is not there, defaults to whatever the bioperl modules have installed,
  57 which itself defaults to SDBM_File.
  58
  59 =head1 USING IT YOURSELF
  60
  61 bp_index.pl is a script that drives the Index modules. If you want to
  62 use this script heavily in your work, if it is Perl based, it is
  63 almost certainly better to look at the code in this script and copy
  64 it across (probably you will be more likely to want to use the bp_fetch
  65 code).
  66
  67 =head1 EXTENDING IT
  68
  69 bp_index is just a wrapper around James Gilbert's excellent Index modules
  70 found in bioperl
  71
  72 =head1 FEEDBACK
  73
  74 =head2 Mailing Lists
  75
  76 User feedback is an integral part of the evolution of this and other
  77 Bioperl modules. Send your comments and suggestions preferably to
  78 the Bioperl mailing list.  Your participation is much appreciated.
  79
  80   bioperl-l@bioperl.org                  - General discussion
  81   http://bioperl.org/wiki/Mailing_lists  - About the mailing lists
  82
  83 =head2 Reporting Bugs
  84
  85 Report bugs to the Bioperl bug tracking system to help us keep track
  86 of the bugs and their resolution. Bug reports can be submitted via the
  87 web:
  88
  89   https://github.com/bioperl/bioperl-live/issues
  90
  91 =head1 AUTHOR - Ewan Birney
  92
  93 Ewan Birney E<lt>birney@ebi.ac.ukE<gt>
  94
  95 =cut
  96
  97 #'
  98 use strict;
  99 use warnings;
 100
 101 #
 102 # Doofus catcher for people who are trying this script without
 103 # installing bioperl
 104 #
 105
 106 use Bio::Index::Fasta;
 107 use Bio::Index::EMBL;
 108 use Bio::Index::Swissprot;
 109 use Bio::Index::GenBank;
 110 use Bio::Index::SwissPfam;
 111
 112 my $dir = $ENV{'BIOPERL_INDEX'};
 113 my $type = $ENV{'BIOPERL_INDEX_TYPE'};
 114 my $fmt = 'Fasta';
 115 my $verbose = 0;
 116
 117 use Getopt::Long;
 118 &GetOptions("f|fmt=s" => \$fmt,
 119             "d|dir=s" => \$dir,
 120             "t|type=s" => \$type,
 121             "v!" => \$verbose);
 122
 123 exec('perldoc',$0) unless @ARGV;
 124
 125 my $name = shift;
 126
 127 if( !$dir ) {
 128     print STDERR "\nNo directory specified for index\nDirectory must be specified by the environment variable BIOPERL_INDEX or -dir option\ngo bp_index with no arguments for more help\n\n";
 129     exit(1);
 130 }
 131
 132 #
 133 # Reset the type if needed
 134 #
 135
 136 if( $type ) {
 137    $Bio::Index::Abstract::USE_DBM_TYPE = $type;
 138 }
 139 #
 140 # Rock and roll...
 141 #
 142 my $index;
 143 $_ = $fmt;
 144 SWITCH : {
 145     /Fasta/i && do {
 146         $index = Bio::Index::Fasta->new("$dir/$name", 'WRITE');
 147         last;
 148     };
 149     /EMBL/i && do {
 150         $index = Bio::Index::EMBL->new("$dir/$name", 'WRITE');
 151         last;
 152     };
 153     /swisspfam|pfam/i && do {
 154         $index = Bio::Index::SwissPfam->new("$dir/$name", 'WRITE');
 155         last;
 156     };
 157     /swiss/i && do {
 158         $index = Bio::Index::Swissprot->new("$dir/$name", 'WRITE');
 159         last;
 160     };
 161     /GenBank/i && do {
 162         $index = Bio::Index::GenBank->new("$dir/$name", 'WRITE');
 163         last;
 164     };
 165     die("No index format called $fmt");
 166 }
 167
 168 if( $verbose != 0 ) {
 169   $index->verbose(1);
 170 }
 171
 172 $index->make_index(@ARGV);
 173
 174 # finished. Neat eh.
 175
 176 #
 177 # if you are using this in a script, to
 178 # to force deallocation + closing of the index, go
 179 # $index = undef;
 180 #
 181
 182
 183
 184
 185
 186
 187