SliceAdaptor.pm 46.5 KB
Newer Older
1 2

#
Simon Potter's avatar
pod  
Simon Potter committed
3
# Ensembl module for Bio::EnsEMBL::DBSQL::SliceAdaptor
4 5
#
#
Graham McVicker's avatar
Graham McVicker committed
6
# Copyright (c) 2004 Ensembl
7 8 9 10 11 12 13
#
# You may distribute this module under the same terms as perl itself

# POD documentation - main docs before the code

=head1 NAME

14 15
Bio::EnsEMBL::DBSQL::SliceAdaptor - A database aware adaptor responsible for
the creation of Slice objects.
16 17 18

=head1 SYNOPSIS

Graham McVicker's avatar
Graham McVicker committed
19 20
  use Bio::EnsEMBL::DBSQL::DBAdaptor;
  use Bio::EnsEMBL::Utils::Slice qw(split_Slices);
21

Graham McVicker's avatar
Graham McVicker committed
22
  $db = Bio::EnsEMBL::DBSQL::DBAdaptor->new(...);
23

Graham McVicker's avatar
Graham McVicker committed
24
  $slice_adaptor = $db->get_SliceAdaptor();
25

Graham McVicker's avatar
Graham McVicker committed
26 27 28 29 30
  # get a slice on the entire chromosome X
  $chr_slice = $slice_adaptor->fetch_by_region('chromosome','X');

  # get a slice for each clone in the database
  foreach $cln_slice (@{$slice_adaptor->fetch_all('clone')}) {
31 32 33
    #do something with clone
  }

Graham McVicker's avatar
Graham McVicker committed
34 35
  # get a slice which is part of NT_004321
  $spctg_slice = $slice_adaptor->fetch_by_region('supercontig','NT_004321',
36
                                                    200_000, 600_000);
37 38


Graham McVicker's avatar
Graham McVicker committed
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69
  # get all non-redundant slices from the highest possible coordinate systems
  $slices = $slice_adaptor->fetch_all('toplevel');

  # include non-reference regions
  $slices = $slice_adaptor->fetch_all('toplevel',undef,1);

  # include non-duplicate regions
  $slices = $slice_adaptor->fetch_all('toplevel', undef, 0, 1);

  # split up a list of slices into smaller slices
  $overlap = 1000;
  $max_length = 1e6;
  $slices = split_Slices($slices, $max_length, $overlap);


  # store a list of slice names in a file
  open(FILE, ">$filename") or die("Could not open file $filename");
  foreach my $slice (@$slices) {
    print FILE $slice->name(), "\n";
  }
  close FILE;

  # retreive a list of slices from a file
  open(FILE, $filename) or die("Could not open file $filename");
  while($name = <FILE>) {
    chomp($name);
    $slice = $slice_adaptor->fetch_by_name($name);

    # do something with slice
  }

70 71
=head1 DESCRIPTION

72
This module is responsible for fetching Slices representing genomic regions
Graham McVicker's avatar
Graham McVicker committed
73
from a database.  A Details on how slices can be used are in the
74
Bio::EnsEMBL::Slice module.
75

76
=head1 CONTACT
77

Graham McVicker's avatar
Graham McVicker committed
78 79 80
This module is part of the Ensembl project http://www.ensembl.org

For more information email <ensembl-dev@ebi.ac.uk>
81

82
=head1 METHODS
83 84 85 86

=cut


87
package Bio::EnsEMBL::DBSQL::SliceAdaptor;
88 89 90
use vars qw(@ISA);
use strict;

91

92 93
use Bio::EnsEMBL::DBSQL::BaseAdaptor;
use Bio::EnsEMBL::Slice;
94
use Bio::EnsEMBL::Mapper;
95

96 97
use Bio::EnsEMBL::Utils::Exception qw(throw deprecate warning);
use Bio::EnsEMBL::Utils::Cache; #CPAN LRU cache
98
use Bio::EnsEMBL::Utils::SeqRegionCache;
99

100

101
@ISA = ('Bio::EnsEMBL::DBSQL::BaseAdaptor');
102

103 104
sub new {
  my $caller = shift;
105

106
  my $class = ref($caller) || $caller;
107

108
  my $self = $class->SUPER::new(@_);
109

110 111 112 113 114 115 116

  return $self;
}


=head2 fetch_by_region

117
  Arg [1]    : string $coord_system_name (optional)
118
               The name of the coordinate system of the slice to be created
119
               This may be a name of an actual coordinate system or an alias
120 121 122 123
               to a coordinate system.  Valid aliases are 'seqlevel' or
               'toplevel'.
  Arg [2]    : string $seq_region_name
               The name of the sequence region that the slice will be
Graham McVicker's avatar
Graham McVicker committed
124
               created on.
125 126 127 128 129 130 131 132
  Arg [3]    : int $start (optional, default = 1)
               The start of the slice on the sequence region
  Arg [4]    : int $end (optional, default = seq_region length)
               The end of the slice on the sequence region
  Arg [5]    : int $strand (optional, default = 1)
               The orientation of the slice on the sequence region
  Arg [6]    : string $version (optional, default = default version)
               The version of the coordinate system to use (e.g. NCBI33)
133 134 135
  Example    : $slice = $slice_adaptor->fetch_by_region('chromosome', 'X');
               $slice = $slice_adaptor->fetch_by_region('clone', 'AC008066.4');
  Description: Retrieves a slice on the requested region.  At a minimum the
136 137 138 139 140 141 142 143 144 145 146
               name the name of the seq_region to fetch must be provided.

               If no coordinate system name is provided than a slice on the
               highest ranked coordinate system with a matching
               seq_region_name will be returned.  If a version but no
               coordinate system name is provided, the same behaviour will
               apply, but only coordinate systems of the appropriate version
               are considered.  The same applies if the 'toplevel' coordinate
               system is specified, however in this case the version is
               ignored.  The coordinate system should always be specified if
               it is known, since this is unambiguous and faster.
147 148 149 150 151 152 153 154 155 156 157 158 159 160

               Some fuzzy matching is performed if no exact match for
               the provided name is found.  This allows clones to be
               fetched even when their version is not known.  For
               example fetch_by_region('clone', 'AC008066') will
               retrieve the sequence_region with name 'AC008066.4'.

               If the requested seq_region is not found in the database undef
               is returned.

  Returntype : Bio::EnsEMBL::Slice or undef
  Exceptions : throw if no seq_region_name is provided
               throw if invalid coord_system_name is provided
               throw if start > end is provided
161
  Caller     : general
162 163 164

=cut

165 166 167 168

#
# ARNE: This subroutine needs simplification!! 
#
169 170 171 172
sub fetch_by_region {
  my ($self, $coord_system_name, $seq_region_name,
      $start, $end, $strand, $version) = @_;

173 174 175 176
  $start  = 1 if(!defined($start));
  $strand = 1 if(!defined($strand));

  throw('seq_region_name argument is required') if(!defined($seq_region_name));
177

178 179
  my $cs;
  my $csa = $self->db->get_CoordSystemAdaptor();
180

181 182 183 184 185 186 187 188 189 190 191 192 193
  if($coord_system_name) {
    $cs = $csa->fetch_by_name($coord_system_name,$version);

    if(!$cs) {
      throw("Unknown coordinate system:\n name='$coord_system_name' " .
            "version='$version'\n");
    }

    # fetching by toplevel is same as fetching w/o name or version
    if($cs->is_top_level()) {
      $cs = undef;
      $version = undef;
    }
194
  }
195 196 197 198 199 200 201 202 203 204 205 206 207

  my $constraint;
  my $sql;
  my @bind_vals;
  my $key;

  if($cs) {
    push @bind_vals, $cs->dbID();
    $sql = "SELECT sr.name,sr.seq_region_id, sr.length, " .
           $cs->dbID() ." FROM seq_region sr ";

    $constraint = "sr.coord_system_id = ?";

208
    $key = "$seq_region_name:".$cs->dbID();
209 210 211 212 213 214 215 216 217 218 219
  } else {
    $sql = "SELECT sr.name, sr.seq_region_id, sr.length, " .
           "       cs.coord_system_id " .
           "FROM   seq_region sr, coord_system cs ";

    $constraint = "sr.coord_system_id = cs.coord_system_id ";
    if($version) {
      $constraint .= "AND cs.version = ? ";
      push @bind_vals, $version;
    }
    $constraint .= "ORDER BY cs.rank DESC";
220 221
  }

222 223
  #check the cache so we only go to the db if necessary
  my $length;
224 225 226 227
  my $arr;
  if( $key ) {
    $arr = $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{$key};
  }
228

229 230
  if( $arr ) {
    $length = $arr->[3];
231
  } else {
232 233
    my $sth = $self->prepare($sql . " WHERE sr.name = ? AND " .
                             $constraint);
Graham McVicker's avatar
Graham McVicker committed
234

235 236 237 238
    # Quotes around "$seq_region_name" are needed so that mysql does not
    # treat chromosomes like '6' as an int.  This was doing horrible
    # inexact matches like '6DR51', '6_UN', etc.
    $sth->execute("$seq_region_name", @bind_vals);
239

240 241
    if($sth->rows() == 0) {
      $sth->finish();
Graham McVicker's avatar
Graham McVicker committed
242

243
      #do fuzzy matching, assuming that we are just missing a version on
244
      #the end of the seq_region name
Graham McVicker's avatar
Graham McVicker committed
245

246 247
      $sth = $self->prepare($sql . " WHERE sr.name LIKE ? AND " . $constraint);
      $sth->execute("$seq_region_name.%", @bind_vals);
248 249

      my $prefix_len = length($seq_region_name) + 1;
250 251
      my $high_ver = undef;
      my $high_cs = $cs;
252 253 254 255

      # find the fuzzy-matched seq_region with the highest postfix (which ought
      # to be a version)

256 257 258 259 260
      my ($tmp_name, $id, $tmp_length, $cs_id);
      $sth->bind_columns(\$tmp_name, \$id, \$tmp_length, \$cs_id);

      while($sth->fetch) {
        my $tmp_cs = ($cs) ? $cs : $csa->fetch_by_dbID($cs_id);
261

262
        # cache values for future reference
263 264 265 266 267
        my $arr = [ $id, $tmp_name, $cs_id, $tmp_length ];
        $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$tmp_name:$cs_id"} =
            $arr;
        $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{"$id"} =
            $arr;
Graham McVicker's avatar
Graham McVicker committed
268

269
        my $tmp_ver = substr($tmp_name, $prefix_len);
270

271 272
        # skip versions which are non-numeric and apparently not versions
        next if($tmp_ver !~ /^\d+$/);
273

274 275 276 277 278 279 280 281
        # take version with highest num, if two versions match take one with
        # highest ranked coord system (lowest num)
        if(!defined($high_ver) || $tmp_ver > $high_ver ||
           ($tmp_ver == $high_ver && $tmp_cs->rank < $high_cs->rank)) {
            $seq_region_name = $tmp_name;
            $length          = $tmp_length;
            $high_ver        = $tmp_ver;
            $high_cs         = $tmp_cs;
Graham McVicker's avatar
Graham McVicker committed
282 283
        }
      }
284 285
      $sth->finish();

286 287
      $cs = $high_cs;

288
      #return if we did not find any appropriate match:
289
      return undef if(!defined($high_ver));
290 291
    } else {

292 293
      my ($id, $cs_id);
      ($seq_region_name, $id, $length, $cs_id) = $sth->fetchrow_array();
294
      $sth->finish();
295 296 297 298 299 300 301 302 303
      
      # cahce to speed up for future queries
      my $arr = [ $id, $seq_region_name, $cs_id, $length ];
      $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$seq_region_name:$cs_id"} =
          $arr;
      $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{"$id"} =
          $arr;

      $cs = $csa->fetch_by_dbID( $cs_id );
304
    }
305
  }
306

307
  $end = $length if(!defined($end));
Graham McVicker's avatar
Graham McVicker committed
308

309 310 311 312
  if($end < $start) {
    throw('start [$start] must be less than or equal to end [$end]');
  }

313
  return Bio::EnsEMBL::Slice->new(-COORD_SYSTEM      => $cs,
314 315 316 317 318 319
                                  -SEQ_REGION_NAME   => $seq_region_name,
                                  -SEQ_REGION_LENGTH => $length,
                                  -START             => $start,
                                  -END               => $end,
                                  -STRAND            => $strand,
                                  -ADAPTOR           => $self);
320 321 322 323
}



324

325
=head2 fetch_by_name
326

Graham McVicker's avatar
Graham McVicker committed
327
  Arg [1]    : string $name
328 329 330 331 332 333 334 335 336
  Example    : $name  = 'chromosome:NCBI34:X:1000000:2000000:1';
               $slice = $slice_adaptor->fetch_by_name($name);
               $slice2 = $slice_adaptor->fetch_by_name($slice3->name());
  Description: Fetches a slice using a slice name (i.e. the value returned by
               the Slice::name method).  This is useful if you wish to 
               store a unique identifier for a slice in a file or database or
               pass a slice over a network.
               Slice::name allows you to serialise/marshall a slice and this
               method allows you to deserialise/unmarshal it.
Graham McVicker's avatar
Graham McVicker committed
337

338 339 340 341
               Returns undef if no seq_region with the provided name exists in
               the database.

  Returntype : Bio::EnsEMBL::Slice or undef
342 343
  Exceptions : throw if incorrent arg provided
  Caller     : Pipeline
344 345 346

=cut

347 348 349
sub fetch_by_name {
  my $self = shift;
  my $name = shift;
Graham McVicker's avatar
Graham McVicker committed
350

351 352
  if(!$name) {
    throw("name argument is required");
Graham McVicker's avatar
Graham McVicker committed
353
  }
354 355 356 357 358

  my @array = split(/:/,$name);

  if(@array != 6) {
    throw("Malformed slice name [$name].  Format is " .
Laura Clarke's avatar
 
Laura Clarke committed
359
        "coord_system:version:name:start:end:strand");
360 361 362 363 364
  }

  my ($cs_name, $cs_version, $seq_region, $start, $end, $strand) = @array;

  return $self->fetch_by_region($cs_name,$seq_region, $start,
365
                                $end, $strand, $cs_version);
Graham McVicker's avatar
Graham McVicker committed
366
}
Graham McVicker's avatar
Graham McVicker committed
367 368 369



370 371 372 373 374 375 376 377
=head2 fetch_by_seq_region_id

  Arg [1]    : string $seq_region_id
               The internal identifier of the seq_region to create this slice
               on
  Example    : $slice = $slice_adaptor->fetch_by_seq_region_id(34413);
  Description: Creates a slice object of an entire seq_region using the
               seq_region internal identifier to resolve the seq_region.
378 379
               Returns undef if no such slice exists.
  Returntype : Bio::EnsEMBL::Slice or undef
Graham McVicker's avatar
Graham McVicker committed
380 381
  Exceptions : none
  Caller     : general
Graham McVicker's avatar
Graham McVicker committed
382 383 384

=cut

385 386 387
sub fetch_by_seq_region_id {
  my ($self, $seq_region_id) = @_;

388
  my $arr = $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{ $seq_region_id };
389 390
  my ($name, $length, $cs);

391 392 393 394
  if( $arr ) {
    my $cs_id;
    ($name, $cs_id, $length ) = ( $arr->[1], $arr->[2], $arr->[3] );
    $cs = $self->db->get_CoordSystemAdaptor->fetch_by_dbID($cs_id);
395 396 397 398 399 400 401
  } else {
    my $sth = $self->prepare("SELECT name, length, coord_system_id " .
                             "FROM seq_region " .
                             "WHERE seq_region_id = ?");

    $sth->execute($seq_region_id);

402
    return undef if($sth->rows() == 0);
403 404 405 406 407 408 409 410

    my $cs_id;
    ($name, $length, $cs_id) = $sth->fetchrow_array();
    $sth->finish();

    $cs = $self->db->get_CoordSystemAdaptor->fetch_by_dbID($cs_id);

    #cache results to speed up repeated queries
411 412 413 414 415 416
    my $arr = [ $seq_region_id, $name, $cs_id, $length ];
      
    $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$name:$cs_id"} =
        $arr;
    $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{"$seq_region_id"} =
        $arr;
417 418
  }

419 420 421 422 423 424 425
  return Bio::EnsEMBL::Slice->new(-COORD_SYSTEM      => $cs,
                                  -SEQ_REGION_NAME   => $name,
                                  -SEQ_REGION_LENGTH => $length,
                                  -START             => 1,
                                  -END               => $length,
                                  -STRAND            => 1,
                                  -ADAPTOR           => $self);
Arne Stabenau's avatar
Arne Stabenau committed
426 427
}

Graham McVicker's avatar
Graham McVicker committed
428

429 430

=head2 get_seq_region_id
Graham McVicker's avatar
Graham McVicker committed
431

Arne Stabenau's avatar
Arne Stabenau committed
432
  Arg [1]    : Bio::EnsEMBL::Slice $slice
433 434 435 436 437 438 439 440 441 442
               The slice to fetch a seq_region_id for
  Example    : $srid = $slice_adaptor->get_seq_region_id($slice);
  Description: Retrieves the seq_region id (in this database) given a slice
               Seq region ids are not stored on the slices themselves
               because they are intended to be somewhat database independant
               and seq_region_ids vary accross databases.
  Returntype : int
  Exceptions : throw if the seq_region of the slice is not in the db
               throw if incorrect arg provided
  Caller     : BaseFeatureAdaptor
Arne Stabenau's avatar
Arne Stabenau committed
443 444

=cut
Graham McVicker's avatar
Graham McVicker committed
445

446 447 448
sub get_seq_region_id {
  my $self = shift;
  my $slice = shift;
Graham McVicker's avatar
Graham McVicker committed
449

450 451 452
  if(!$slice || !ref($slice) || !$slice->isa('Bio::EnsEMBL::Slice')) {
    throw('Slice argument is required');
  }
453
  
454
  my $seq_region_name = $slice->seq_region_name();
455 456
  my $key = $seq_region_name.":".$slice->coord_system->dbID();
  my $arr = $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$key"};
457

458 459
  if( $arr ) {
    return $arr->[0];
460 461
  }

462
  my $cs_id = $slice->coord_system->dbID();
463 464 465 466 467 468

  my $sth = $self->prepare("SELECT seq_region_id, length " .
                           "FROM seq_region " .
                           "WHERE name = ? AND coord_system_id = ?");

  #force seq_region_name cast to string so mysql cannot treat as int
469
  $sth->execute("$seq_region_name", $cs_id );
470 471 472

  if($sth->rows() != 1) {
    throw("Non existant or ambigous seq_region:\n" .
473 474 475
          "  coord_system=[$cs_id],\n" .
          "  name=[$seq_region_name],\n");

476 477 478 479 480 481
  }

  my($seq_region_id, $length) = $sth->fetchrow_array();
  $sth->finish();

  #cache information for future requests
482 483 484 485 486 487 488
  $arr = [ $seq_region_id, $seq_region_name, $cs_id, $length ];
  
  $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$seq_region_name:$cs_id"} =
      $arr;
  $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{"$seq_region_id"} =
      $arr;
  
489
  return $seq_region_id;
Arne Stabenau's avatar
Arne Stabenau committed
490
}
491 492


Arne Stabenau's avatar
Arne Stabenau committed
493

494 495 496 497 498 499 500 501 502
=head2 fetch_all

  Arg [1]    : string $coord_system_name
               The name of the coordinate system to retrieve slices of.
               This may be a name of an acutal coordinate system or an alias
               to a coordinate system.  Valid aliases are 'seqlevel' or
               'toplevel'.
  Arg [2]    : string $coord_system_version (optional)
               The version of the coordinate system to retrieve slices of
503
  Arg [3]    : bool $include_non_reference (optional)
504 505 506
               If this argument is not provided then only reference slices
               will be returned. If set, both reference and non refeference
               slices will be rerurned.
Graham McVicker's avatar
Graham McVicker committed
507 508
  Arg [4]    : int $include_duplicates (optional)
               If set duplicate regions will be returned.
509

510 511 512
  Example    : @chromos = @{$slice_adaptor->fetch_all('chromosome','NCBI33')};
               @contigs = @{$slice_adaptor->fetch_all('contig')};

Graham McVicker's avatar
Graham McVicker committed
513 514 515 516 517
               # get even non-reference regions
               @slices = @{$slice_adaptor->fetch_all('toplevel',undef,1)};

               # include duplicate regions (such as pseudo autosomal regions)
               @slices = @{$slice_adaptor->fetch_all('toplevel', undef,0,1)};
518 519 520 521 522 523

  Description: Retrieves slices of all seq_regions for a given coordinate
               system.  This is analagous to the methods fetch_all which were
               formerly on the ChromosomeAdaptor, RawContigAdaptor and
               CloneAdaptor classes.  Slices fetched span the entire
               seq_regions and are on the forward strand.
524 525
               If the coordinate system with the provided name and version
               does not exist an empty list is returned.
526
               If the coordinate system name provided is 'toplevel', all
527
               non-redundant toplevel slices are returned.
528 529 530 531

               Retrieved slices can be broken into smaller slices using the
               Bio::EnsEMBL::Utils::Slice module.

532
  Returntype : listref of Bio::EnsEMBL::Slices
533
  Exceptions : none
534
  Caller     : general
Web Admin's avatar
Web Admin committed
535 536 537

=cut

538 539 540 541 542
sub fetch_all {
  my $self = shift;
  my $cs_name = shift;
  my $cs_version = shift || '';

543
  my ($include_non_reference, $include_duplicates) = @_;
Web Admin's avatar
Web Admin committed
544

545 546 547
  #
  # verify existance of requested coord system and get its id
  #
548 549 550 551 552 553
  my $csa       = $self->db->get_CoordSystemAdaptor();
  my $orig_cs   = $csa->fetch_by_name($cs_name, $cs_version);

  return [] if(!$orig_cs);

  my $sth;
554

555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573
  my %bad_vals=();
  #
  # Get a hash of non reference seq regions
  #
  if(!$include_non_reference){
    my $sth2 = $self->prepare(
			      "SELECT sr.seq_region_id ".
			      "FROM seq_region sr, seq_region_attrib sra, attrib_type at ".
			      " WHERE at.code='non_ref'".
			      "  AND sra.seq_region_id=sr.seq_region_id ".
			      "  AND at.attrib_type_id=sra.attrib_type_id " );
    $sth2->execute();
    my ($seq_region_id);
    $sth2->bind_columns(\$seq_region_id);
    while($sth2->fetch()) {
      $bad_vals{$seq_region_id} = 1;
    }
    $sth2->finish();
  }
574
  #
575
  # Retrieve the seq_regions from the database
576
  #
577
  if($orig_cs->is_top_level()) {
578 579 580 581 582 583 584 585
   $sth =
       $self->prepare("SELECT sr.seq_region_id, sr.name, sr.length, " .
                      "       sr.coord_system_id " .
                      "FROM seq_region sr, " .
                      "     seq_region_attrib sra, attrib_type at " .
                      "WHERE at.code='toplevel' " .
                      "AND   at.attrib_type_id=sra.attrib_type_id " .
                      "AND   sra.seq_region_id=sr.seq_region_id");
586 587
    $sth->execute();
  } else {
588 589 590 591 592
     $sth =
       $self->prepare('SELECT seq_region_id, name, length, coord_system_id ' .
                      'FROM   seq_region ' .
                      'WHERE  coord_system_id =?');
     $sth->execute($orig_cs->dbID);
593
  }
594

595 596
  my ($seq_region_id, $name, $length, $cs_id);
  $sth->bind_columns(\$seq_region_id, \$name, \$length, \$cs_id);
597

598
  my $cache_count = 0;
599 600 601

  my @out;
  while($sth->fetch()) {
602 603
    if(!defined($bad_vals{$seq_region_id})){
      my $cs = $csa->fetch_by_dbID($cs_id);
604

605
      if(!$cs) {
606
        throw("seq_region $name references non-existent coord_system $cs_id.");
607
      }
608

609 610
      #cache values for future reference, but stop adding to the cache once we
      #we know we have filled it up
611
      if($cache_count < $Bio::EnsEMBL::Utils::SeqRegionCache::SEQ_REGION_CACHE_SIZE) {
612 613 614 615 616 617 618
        my $arr = [ $seq_region_id, $name, $cs_id, $length ];
        
        $Bio::EnsEMBL::Utils::SeqRegionCache::sr_name_cache{"$name:$cs_id"} =
            $arr;
        $Bio::EnsEMBL::Utils::SeqRegionCache::sr_id_cache{"$seq_region_id"} =
            $arr;

619
        $cache_count++;
620
      }
621 622 623 624 625 626 627 628 629 630 631

      my $slice = Bio::EnsEMBL::Slice->new
        (-START             => 1,
         -END               => $length,
         -STRAND            => 1,
         -SEQ_REGION_NAME   => $name,
         -SEQ_REGION_LENGTH => $length,
         -COORD_SYSTEM      => $cs,
         -ADAPTOR           => $self);

      if(!defined($include_duplicates) or !$include_duplicates){
632 633 634 635 636 637 638 639 640 641 642 643
        # test if this slice *could* have a duplicate (exception) region
        $self->_build_exception_cache() if(!exists $self->{'asm_exc_cache'});
        if(exists $self->{asm_exc_cache}->{$seq_region_id}) {

          # Dereference symlinked assembly regions.  Take out
          # any regions which are symlinked because these are duplicates
          my @projection = @{$self->fetch_normalized_slice_projection($slice)};
          foreach my $segment ( @projection) {
            if($segment->[2]->seq_region_name() eq $slice->seq_region_name() &&
               $segment->[2]->coord_system->equals($slice->coord_system)) {
              push @out, $segment->[2];
            }
644
          }
645 646 647
        } else {
          # no duplicate regions
          push @out, $slice;
648
        }
649 650
      } else {
        # we want duplicates anyway so do not do any checks
651
        push @out, $slice;
652
      }
653
    }
654
  }
655
  $sth->finish();
656

657
  return \@out;
Web Admin's avatar
Web Admin committed
658 659 660
}


661

Web Admin's avatar
Web Admin committed
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691
=head2 fetch_by_band

 Title   : fetch_by_band
 Usage   :
 Function: create a Slice representing a series of bands
 Example :
 Returns :
 Args    : the band name

=cut

sub fetch_by_band {
  my ($self,$band) = @_;

  my $sth = $self->db->prepare
        ("select s.name,max(k.seq_region_id)-min(k.seq_region_id, min(k.seq_region_start), max(k.seq_region_id) " .
         "from karyotype as k " .
         "where k.band like ? and k.seq_region_id = s.seq_region_id");

  $sth->execute( "$band%" );
  my ( $seq_region_name, $discrepancy, $seq_region_start, $seq_region_end) = $sth->fetchrow_array;

  if($seq_region_name && $discrepancy>0) {
    throw("Band maps to multiple seq_regions");
  } else {
    return $self->fetch_by_region('toplevel',$seq_region_name,$seq_region_start,$seq_region_end);
  }
  throw("Band not recognised in database");
}

692 693 694 695 696 697 698 699 700 701 702 703 704 705
=head2 fetch_by_chr_band

 Title   : fetch_by_chr_band
 Usage   :
 Function: create a Slice representing a series of bands
 Example :
 Returns :
 Args    : the band name

=cut

sub fetch_by_chr_band {
  my ($self,$chr,$band) = @_;

Web Admin's avatar
Web Admin committed
706
  my $chr_slice = $self->fetch_by_region('toplevel', $chr);
707 708 709
  my $seq_region_id = $self->get_seq_region_id($chr_slice);

  my $sth = $self->db->prepare
Web Admin's avatar
Web Admin committed
710
        ("select min(k.seq_region_start), max(k.seq_region_end) " .
711 712 713 714 715 716 717
         "from karyotype as k " .
         "where k.seq_region_id = ? and k.band like ?");

  $sth->execute( $seq_region_id, "$band%" );
  my ( $slice_start, $slice_end) = $sth->fetchrow_array;

  if(defined $slice_start) {
Web Admin's avatar
Web Admin committed
718
    return $self->fetch_by_region('toplevel',$chr,$slice_start,$slice_end);
719 720 721
  }

  throw("Band not recognised in database");
722 723 724 725
}



Web Admin's avatar
Web Admin committed
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
=head2 fetch_by_exon_stable_id

  Arg [1]    : string $exonid
               The stable id of the exon around which the slice is 
               desired
  Arg [2]    : (optional) int $size
               The length of the flanking regions the slice should encompass 
               on either side of the exon (0 by default)
  Example    : $slc = $sa->fetch_by_exon_stable_id('ENSE00000302930',10);
  Description: Creates a slice around the region of the specified exon. 
               If a context size is given, the slice is extended by that 
               number of basepairs on either side of the
               transcript.
  Returntype : Bio::EnsEMBL::Slice
  Exceptions : Thrown if the exon is not in the database.
  Caller     : general

=cut

sub fetch_by_exon_stable_id{
  my ($self,$exonid,$size) = @_;

  throw('Exon argument is required.') if(!$exonid);

  my $ea = $self->db->get_ExonAdaptor;
  my $exon = $ea->fetch_by_stable_id($exonid);

  throw("Exon [$exonid] does not exist in DB.") if(!$exon);

  return $self->fetch_by_Feature($exon, $size);
}

Graham McVicker's avatar
Graham McVicker committed
758
=head2 fetch_by_transcript_stable_id
759

Graham McVicker's avatar
Graham McVicker committed
760 761 762 763 764 765 766 767 768
  Arg [1]    : string $transcriptid
               The stable id of the transcript around which the slice is 
               desired
  Arg [2]    : (optional) int $size
               The length of the flanking regions the slice should encompass 
               on either side of the transcript (0 by default)
  Example    : $slc = $sa->fetch_by_transcript_stable_id('ENST00000302930',10);
  Description: Creates a slice around the region of the specified transcript. 
               If a context size is given, the slice is extended by that 
769 770
               number of basepairs on either side of the
               transcript.
Graham McVicker's avatar
Graham McVicker committed
771
  Returntype : Bio::EnsEMBL::Slice
772
  Exceptions : Thrown if the transcript is not in the database.
Graham McVicker's avatar
Graham McVicker committed
773
  Caller     : general
774 775 776

=cut

Graham McVicker's avatar
Graham McVicker committed
777
sub fetch_by_transcript_stable_id{
778 779
  my ($self,$transcriptid,$size) = @_;

780 781
  throw('Transcript argument is required.') if(!$transcriptid);

782
  my $ta = $self->db->get_TranscriptAdaptor;
783
  my $transcript = $ta->fetch_by_stable_id($transcriptid);
784

785 786 787 788
  throw("Transcript [$transcriptid] does not exist in DB.") if(!$transcript);

  return $self->fetch_by_Feature($transcript, $size);
}
789

Graham McVicker's avatar
Graham McVicker committed
790 791


Graham McVicker's avatar
Graham McVicker committed
792 793
=head2 fetch_by_transcript_id

Graham McVicker's avatar
Graham McVicker committed
794 795 796 797 798 799 800
  Arg [1]    : int $transcriptid
               The unique database identifier of the transcript around which 
               the slice is desired
  Arg [2]    : (optional) int $size
               The length of the flanking regions the slice should encompass 
               on either side of the transcript (0 by default)
  Example    : $slc = $sa->fetch_by_transcript_id(24, 1000);
801 802 803 804
  Description: Creates a slice around the region of the specified transcript.
               If a context size is given, the slice is extended by that
               number of basepairs on either side of the
               transcript.
Graham McVicker's avatar
Graham McVicker committed
805
  Returntype : Bio::EnsEMBL::Slice
806 807
  Exceptions : throw on incorrect args
               throw if transcript is not in database
Graham McVicker's avatar
Graham McVicker committed
808
  Caller     : general
809 810 811

=cut

Graham McVicker's avatar
Graham McVicker committed
812
sub fetch_by_transcript_id {
813
  my ($self,$transcriptid,$size) = @_;
Graham McVicker's avatar
Graham McVicker committed
814

815
  throw('Transcript id argument is required.') if(!$transcriptid);
Graham McVicker's avatar
Graham McVicker committed
816

817 818 819 820 821 822
  my $transcript_adaptor = $self->db()->get_TranscriptAdaptor();
  my $transcript = $transcript_adaptor->fetch_by_dbID($transcriptid);

  throw("Transcript [$transcriptid] does not exist in DB.") if(!$transcript);

  return $self->fetch_by_Feature($transcript, $size);
823 824
}

825 826


Graham McVicker's avatar
Graham McVicker committed
827
=head2 fetch_by_gene_stable_id
828

Graham McVicker's avatar
Graham McVicker committed
829
  Arg [1]    : string $geneid
830
               The stable id of the gene around which the slice is
Graham McVicker's avatar
Graham McVicker committed
831 832
               desired
  Arg [2]    : (optional) int $size
Graham McVicker's avatar
Graham McVicker committed
833
               The length of the flanking regions the slice should encompass
Graham McVicker's avatar
Graham McVicker committed
834 835
               on either side of the gene (0 by default)
  Example    : $slc = $sa->fetch_by_transcript_stable_id('ENSG00000012123',10);
Graham McVicker's avatar
Graham McVicker committed
836 837 838
  Description: Creates a slice around the region of the specified gene.
               If a context size is given, the slice is extended by that
               number of basepairs on either side of the gene.
839
               The slice will be created in the genes native coordinate system.
Graham McVicker's avatar
Graham McVicker committed
840
  Returntype : Bio::EnsEMBL::Slice
841 842
  Exceptions : throw on incorrect args
               throw if transcript does not exist
Graham McVicker's avatar
Graham McVicker committed
843
  Caller     : general
844 845 846

=cut

847 848
sub fetch_by_gene_stable_id {
  my ($self,$geneid,$size) = @_;
849

850
  throw('Gene argument is required.') if(!$geneid);
851

852 853
  my $gene_adaptor = $self->db->get_GeneAdaptor();
  my $gene = $gene_adaptor->fetch_by_stable_id($geneid);
854

855
  throw("Gene [$geneid] does not exist in DB.") if(!$gene);
856

857
  return $self->fetch_by_Feature($gene, $size);
858 859 860
}


861

862
=head2 fetch_by_Feature
Graham McVicker's avatar
Graham McVicker committed
863

864 865 866 867
  Arg [1]    : Bio::EnsEMBL::Feature $feat
               The feature to fetch the slice around
  Arg [2]    : int size (optional)
               The desired number of flanking basepairs around the feature.
868 869
               The size may also be provided as a percentage of the feature 
               size such as 200% or 80.5%.
870 871 872 873 874 875
  Example    : $slice = $slice_adaptor->fetch_by_Feature($feat, 100);
  Description: Retrieves a slice around a specific feature.  All this really
               does is return a resized version of the slice that the feature
               is already on. Note that slices returned from this method
               are always on the forward strand of the seq_region regardless of
               the strandedness of the feature passed in.
Graham McVicker's avatar
Graham McVicker committed
876
  Returntype : Bio::EnsEMBL::Slice
877 878 879 880
  Exceptions : throw if the feature does not have an attached slice
               throw if feature argument is not provided
  Caller     : fetch_by_gene_stable_id, fetch_by_transcript_stable_id,
               fetch_by_gene_id, fetch_by_transcript_id
Graham McVicker's avatar
Graham McVicker committed
881 882 883

=cut

884 885
sub fetch_by_Feature{
  my ($self, $feature, $size) = @_;
Graham McVicker's avatar
Graham McVicker committed
886

887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918
  $size ||= 0;

  if(!ref($feature) || !$feature->isa('Bio::EnsEMBL::Feature')) {
    throw('Feature argument expected.');
  }

  my $slice = $feature->slice();
  if(!$slice || !$slice->isa('Bio::EnsEMBL::Slice')) {
    throw('Feature must be attached to a valid slice.');
  }

  my $fstart = $feature->start();
  my $fend   = $feature->end();
  if(!defined($fstart) || !defined($fend)) {
    throw('Feature must have defined start and end.');
  }

  #convert the feature slice coordinates to seq_region coordinates
  my $slice_start  = $slice->start();
  my $slice_end    = $slice->end();
  my $slice_strand = $slice->strand();
  if($slice_start != 1 || $slice_strand != 1) {
    if($slice_strand == 1) {
      $fstart = $fstart + $slice_start - 1;
      $fend   = $fend   + $slice_start - 1;
    } else {
      my $tmp_start = $fstart;
      $fstart = $slice_end - $fend      + 1;
      $fend   = $slice_end - $tmp_start + 1;
    }
  }

919 920 921 922
  ## Size may be stored as a %age of the length of the feature
  ## Size = 100% gives no context
  ## Size = 200% gives context - 50% the size of the feature either side of 
  ## feature
Web Admin's avatar
Web Admin committed
923 924 925

  $size = int( ($1-100)/200 * ($fend-$fstart+1) ) if( $size =~/([\d+\.]+)%/ );

926
  #return a new slice covering the region of the feature
927 928 929 930 931 932 933 934
  return Bio::EnsEMBL::Slice->new
    (-seq_region_name   => $slice->seq_region_name,
     -seq_region_length => $slice->seq_region_length,
     -coord_system      => $slice->coord_system,
     -start             => $fstart - $size,
     -end               => $fend + $size,
     -strand            => 1,
     -adaptor           => $self);
935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966
}



=head2 fetch_by_misc_feature_attribute

  Arg [1]    : string $attribute_type
               The code of the attribute type
  Arg [2]    : (optional) string $attribute_value
               The value of the attribute to fetch by
  Arg [3]    : (optional) int $size
               The amount of flanking region around the misc feature desired.
  Example    : $slice = $sa->fetch_by_misc_feature_attribute('superctg',
                                                             'NT_030871');
               $slice = $sa->fetch_by_misc_feature_attribute('synonym',
                                                             'AL00012311',
                                                             $flanking);
  Description: Fetches a slice around a MiscFeature with a particular
               attribute type and value. If no value is specified then
               the feature with the particular attribute is used.
               If no size is specified then 0 is used.
  Returntype : Bio::EnsEMBL::Slice
  Exceptions : Throw if no feature with the specified attribute type and value
               exists in the database
               Warning if multiple features with the specified attribute type
               and value exist in the database.
  Caller     : webcode

=cut

sub fetch_by_misc_feature_attribute {
  my ($self, $attrib_type_code, $attrib_value, $size) = @_;
Graham McVicker's avatar
Graham McVicker committed
967

968
  my $mfa = $self->db()->get_MiscFeatureAdaptor();
Web Admin's avatar
Web Admin committed
969

970 971
  my $feats = $mfa->fetch_all_by_attribute_type_value($attrib_type_code,
                                                   $attrib_value);
Web Admin's avatar
Web Admin committed
972

973 974 975 976
  if(@$feats == 0) {
    throw("MiscFeature with $attrib_type_code=$attrib_value does " .
          "not exist in DB.");
  }
Graham McVicker's avatar
Graham McVicker committed
977

978 979 980 981
  if(@$feats > 1) {
    warning("MiscFeature with $attrib_type_code=$attrib_value is " .
            "ambiguous - using first one found.");
  }
982

983
  my ($feat) = @$feats;
984

985
  return $self->fetch_by_Feature($feat, $size);
Graham McVicker's avatar
Graham McVicker committed
986 987
}

988
=head2 fetch_normalized_slice_projection
Graham McVicker's avatar
Graham McVicker committed
989

990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
  Arg [1]    : Bio::EnsEMBL::Slice $slice
  Example    :  ( optional )
  Description: gives back a project style result. The returned slices 
               represent the areas to which there are symlinks for the 
               given slice. start, end show which area on given slice is 
               symlinked
  Returntype : [[start,end,$slice][]]
  Exceptions : none
  Caller     : BaseFeatureAdaptor

=cut


sub fetch_normalized_slice_projection {
  my $self = shift;
  my $slice = shift;

  my $slice_seq_region_id = $self->get_seq_region_id( $slice );

1009
  $self->_build_exception_cache() if(!exists($self->{'asm_exc_cache'}));
1010

1011
  my $result = $self->{'asm_exc_cache'}->{$slice_seq_region_id};
1012

1013
  $result ||= [];
1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024