| 1 |
#!/usr/bin/perl |
| 2 |
|
| 3 |
use strict; |
| 4 |
use warnings; |
| 5 |
use lib '../blib/lib'; |
| 6 |
use Bio::DB::GFF; |
| 7 |
use Getopt::Long; |
| 8 |
|
| 9 |
=head1 NAME |
| 10 |
|
| 11 |
bp_load_gff.pl - Load a Bio::DB::GFF database from GFF files. |
| 12 |
|
| 13 |
=head1 SYNOPSIS |
| 14 |
|
| 15 |
% bp_load_gff.pl -d testdb -u user -p pw |
| 16 |
--dsn 'dbi:mysql:database=dmel_r5_1;host=myhost;port=myport' |
| 17 |
dna1.fa dna2.fa features1.gff features2.gff ... |
| 18 |
|
| 19 |
=head1 DESCRIPTION |
| 20 |
|
| 21 |
This script loads a Bio::DB::GFF database with the features contained |
| 22 |
in a list of GFF files and/or FASTA sequence files. You must use the |
| 23 |
exact variant of GFF described in L<Bio::DB::GFF>. Various |
| 24 |
command-line options allow you to control which database to load and |
| 25 |
whether to allow an existing database to be overwritten. |
| 26 |
|
| 27 |
This script uses the Bio::DB::GFF interface, and so works with all |
| 28 |
database adaptors currently supported by that module (MySQL, Oracle, |
| 29 |
PostgreSQL soon). However, it is slow. For faster loading, see the |
| 30 |
MySQL-specific L<bp_bulk_load_gff.pl> and L<bp_fast_load_gff.pl> scripts. |
| 31 |
|
| 32 |
=head2 NOTES |
| 33 |
|
| 34 |
If the filename is given as "-" then the input is taken from standard |
| 35 |
input. Compressed files (.gz, .Z, .bz2) are automatically |
| 36 |
uncompressed. |
| 37 |
|
| 38 |
FASTA format files are distinguished from GFF files by their filename |
| 39 |
extensions. Files ending in .fa, .fasta, .fast, .seq, .dna and their |
| 40 |
uppercase variants are treated as FASTA files. Everything else is |
| 41 |
treated as a GFF file. If you wish to load -fasta files from STDIN, |
| 42 |
then use the -f command-line swith with an argument of '-', as in |
| 43 |
|
| 44 |
gunzip my_data.fa.gz | bp_fast_load_gff.pl -d test -f - |
| 45 |
|
| 46 |
On the first load of a database, you will see a number of "unknown |
| 47 |
table" errors. This is normal. |
| 48 |
|
| 49 |
About maxfeature: the default value is 100,000,000 bases. If you have |
| 50 |
features that are close to or greater that 100Mb in length, then the |
| 51 |
value of maxfeature should be increased to 1,000,000,000, or another |
| 52 |
power of 10. |
| 53 |
|
| 54 |
=head1 COMMAND-LINE OPTIONS |
| 55 |
|
| 56 |
Command-line options can be abbreviated to single-letter options. |
| 57 |
e.g. -d instead of --database. |
| 58 |
|
| 59 |
--dsn <dsn> Data source (default dbi:mysql:test) |
| 60 |
--adaptor <adaptor> Schema adaptor (default dbi::mysqlopt) |
| 61 |
--user <user> Username for mysql authentication |
| 62 |
--pass <password> Password for mysql authentication |
| 63 |
--fasta <path> Fasta file or directory containing fasta files for the DNA |
| 64 |
--create Force creation and initialization of database |
| 65 |
--maxfeature Set the value of the maximum feature size (default 100 Mb; must be a power of 10) |
| 66 |
--group A list of one or more tag names (comma or space separated) |
| 67 |
to be used for grouping in the 9th column. |
| 68 |
--upgrade Upgrade existing database to current schema |
| 69 |
--gff3_munge Activate GFF3 name munging (see Bio::DB::GFF) |
| 70 |
--quiet No progress reports |
| 71 |
--summary Generate summary statistics for drawing coverage histograms. |
| 72 |
This can be run on a previously loaded database or during |
| 73 |
the load. |
| 74 |
|
| 75 |
=head1 SEE ALSO |
| 76 |
|
| 77 |
L<Bio::DB::GFF>, L<bulk_load_gff.pl>, L<bp_load_gff.pl> |
| 78 |
|
| 79 |
=head1 AUTHOR |
| 80 |
|
| 81 |
Lincoln Stein, [email protected] |
| 82 |
|
| 83 |
Copyright (c) 2002 Cold Spring Harbor Laboratory |
| 84 |
|
| 85 |
This library is free software; you can redistribute it and/or modify |
| 86 |
it under the same terms as Perl itself. See DISCLAIMER.txt for |
| 87 |
disclaimers of warranty. |
| 88 |
|
| 89 |
=cut |
| 90 |
|
| 91 |
my ($DSN,$ADAPTOR,$CREATE,$USER,$PASSWORD,$FASTA,$UPGRADE,$MAX_BIN,$GROUP_TAG,$MUNGE,$QUIET,$SUMMARY_STATS); |
| 92 |
|
| 93 |
GetOptions ('dsn:s' => \$DSN, |
| 94 |
'adaptor:s' => \$ADAPTOR, |
| 95 |
'u|user:s' => \$USER, |
| 96 |
'p|password:s' => \$PASSWORD, |
| 97 |
'fasta:s' => \$FASTA, |
| 98 |
'upgrade' => \$UPGRADE, |
| 99 |
'maxbin|maxfeature:s' => \$MAX_BIN, |
| 100 |
'group:s' => \$GROUP_TAG, |
| 101 |
'gff3_munge' => \$MUNGE, |
| 102 |
'quiet' => \$QUIET, |
| 103 |
'summary' => \$SUMMARY_STATS, |
| 104 |
'create' => \$CREATE) or (system('pod2text',$0), exit -1); |
| 105 |
|
| 106 |
# some local defaults |
| 107 |
$DSN ||= 'dbi:mysql:test'; |
| 108 |
$ADAPTOR ||= 'dbi::mysqlopt'; |
| 109 |
$MAX_BIN ||= 1_000_000_000; # to accomodate human-sized chromosomes |
| 110 |
|
| 111 |
my @args; |
| 112 |
push @args,(-user=>$USER) if defined $USER; |
| 113 |
push @args,(-pass=>$PASSWORD) if defined $PASSWORD; |
| 114 |
push @args,(-preferred_groups=>[split(/[,\s+]+/,$GROUP_TAG)]) if defined $GROUP_TAG; |
| 115 |
push @args,(-create=>1) if $CREATE; |
| 116 |
push @args,(-write=>1); |
| 117 |
|
| 118 |
my $db = Bio::DB::GFF->new(-adaptor=>$ADAPTOR,-dsn => $DSN,@args) |
| 119 |
or die "Can't open database: ",Bio::DB::GFF->error,"\n"; |
| 120 |
|
| 121 |
$db->gff3_name_munging(1) if $MUNGE; |
| 122 |
|
| 123 |
if ($CREATE) { |
| 124 |
$SUMMARY_STATS++; |
| 125 |
$MAX_BIN ? $db->initialize(-erase=>1,-MAX_BIN=>$MAX_BIN) : |
| 126 |
$db->initialize(1); |
| 127 |
} elsif ($UPGRADE) { |
| 128 |
warn qq(expect to see several "table already exists" messages\n); |
| 129 |
$db->initialize(0); |
| 130 |
my $dbi = $db->dbh; # get the raw database handle |
| 131 |
my ($count) = $dbi->selectrow_array('SELECT COUNT(*) FROM fnote'); |
| 132 |
if (defined($count) && $count > 0) { |
| 133 |
warn qq(fnote table detected. Translating into fattribute table. This may take a while.\n); |
| 134 |
$dbi->do("INSERT INTO fattribute VALUES (1,'Note')") or die "failed: ",$dbi->errstr; |
| 135 |
$dbi->do("INSERT INTO fattribute_to_feature (fid,fattribute_id,fattribute_value) SELECT fnote.fid,1,fnote FROM fnote") or die "failed: ",$dbi->errstr; |
| 136 |
warn qq(Schema successfully upgraded. You might want to drop the fnote table when you're sure everything's working.\n); |
| 137 |
} |
| 138 |
} |
| 139 |
|
| 140 |
my (@gff,@fasta); |
| 141 |
foreach (@ARGV) { |
| 142 |
if (/\.(fa|fasta|dna|seq|fast)$/i) { |
| 143 |
push @fasta,$_; |
| 144 |
} else { |
| 145 |
push @gff,$_; |
| 146 |
} |
| 147 |
} |
| 148 |
|
| 149 |
for my $file (@gff) { |
| 150 |
warn "$file: loading...\n"; |
| 151 |
my $loaded = $db->load_gff($file,!$QUIET); |
| 152 |
warn "$file: $loaded records loaded\n"; |
| 153 |
} |
| 154 |
|
| 155 |
unshift @fasta,$FASTA if defined $FASTA; |
| 156 |
|
| 157 |
for my $file (@fasta) { |
| 158 |
warn "Loading fasta ",(-d $file?"directory":"file"), " $file\n"; |
| 159 |
my $loaded = $db->load_fasta($file,!$QUIET); |
| 160 |
warn "$file: $loaded records loaded\n"; |
| 161 |
} |
| 162 |
|
| 163 |
if ($SUMMARY_STATS) { |
| 164 |
warn "Building summary statistics for coverage histograms...\n"; |
| 165 |
$db->build_summary_statistics; |
| 166 |
} |