SliceAdaptor.pm 49.1 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
use Bio::EnsEMBL::Utils::Exception qw(throw deprecate warning);
97

98

99
@ISA = ('Bio::EnsEMBL::DBSQL::BaseAdaptor');
100

101 102
sub new {
  my $caller = shift;
103

104
  my $class = ref($caller) || $caller;
105

106
  my $self = $class->SUPER::new(@_);
107

108 109 110 111 112 113
  # use a cache which is shared and also used by the assembly
  # mapper adaptor
  my $seq_region_cache = $self->db->get_SeqRegionCache();

  $self->{'sr_name_cache'} = $seq_region_cache->{'name_cache'};
  $self->{'sr_id_cache'}   = $seq_region_cache->{'id_cache'};
114 115 116 117 118 119 120

  return $self;
}


=head2 fetch_by_region

121
  Arg [1]    : string $coord_system_name (optional)
122
               The name of the coordinate system of the slice to be created
123
               This may be a name of an actual coordinate system or an alias
124 125 126 127
               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
128
               created on.
129 130 131 132 133 134 135 136
  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)
137 138 139
  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
140 141 142 143 144 145 146 147 148 149 150
               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.
151 152 153 154 155 156 157 158 159 160 161 162 163 164

               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
165
  Caller     : general
166
  Status     : Stable
167 168 169

=cut

170 171 172 173

#
# ARNE: This subroutine needs simplification!! 
#
174 175 176 177
sub fetch_by_region {
  my ($self, $coord_system_name, $seq_region_name,
      $start, $end, $strand, $version) = @_;

178 179 180 181
  $start  = 1 if(!defined($start));
  $strand = 1 if(!defined($strand));

  throw('seq_region_name argument is required') if(!defined($seq_region_name));
182

183 184
  my $cs;
  my $csa = $self->db->get_CoordSystemAdaptor();
185

186 187 188 189 190 191 192 193 194 195 196 197 198
  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;
    }
199
  }
200 201 202 203 204 205 206 207 208 209 210 211 212

  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 = ?";

213
    $key = "$seq_region_name:".$cs->dbID();
214 215 216 217 218 219 220 221 222 223
  } 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;
    }
224
    $constraint .= "ORDER BY cs.rank ASC";
225 226
  }

227 228
  #check the cache so we only go to the db if necessary
  my $length;
229 230
  my $arr;
  if( $key ) {
231
    $arr = $self->{'sr_name_cache'}->{$key};
232
  }
233

234 235
  if( $arr ) {
    $length = $arr->[3];
236
  } else {
237 238
    my $sth = $self->prepare($sql . " WHERE sr.name = ? AND " .
                             $constraint);
Graham McVicker's avatar
Graham McVicker committed
239

240 241 242
    # 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.
243 244 245 246 247
    $sth->bind_param(1,"$seq_region_name",SQL_VARCHAR);
    $sth->bind_param(2,$cs->dbID,SQL_INTEGER) if ($cs);
    $sth->bind_param(2,$version,SQL_VARCHAR)  if ($version);	

    $sth->execute();
248

249 250
    if($sth->rows() == 0) {
      $sth->finish();
Graham McVicker's avatar
Graham McVicker committed
251

252
      #do fuzzy matching, assuming that we are just missing a version on
253
      #the end of the seq_region name
Graham McVicker's avatar
Graham McVicker committed
254

255
      $sth = $self->prepare($sql . " WHERE sr.name LIKE ? AND " . $constraint);
256 257 258 259
      $sth->bind_param(1,"$seq_region_name.%",SQL_VARCHAR);
      $sth->bind_param(2,$cs->dbID,SQL_INTEGER) if ($cs);
      $sth->bind_param(2,$version,SQL_VARCHAR) if ($version);
      $sth->execute();
260 261

      my $prefix_len = length($seq_region_name) + 1;
262 263
      my $high_ver = undef;
      my $high_cs = $cs;
264 265 266 267

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

268 269 270 271 272
      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);
273

274
        # cache values for future reference
275
        my $arr = [ $id, $tmp_name, $cs_id, $tmp_length ];
276 277
        $self->{'sr_name_cache'}->{"$tmp_name:$cs_id"} = $arr;
        $self->{'sr_id_cache'}->{"$id"} = $arr;
Graham McVicker's avatar
Graham McVicker committed
278

279
        my $tmp_ver = substr($tmp_name, $prefix_len);
280

281 282
        # skip versions which are non-numeric and apparently not versions
        next if($tmp_ver !~ /^\d+$/);
283

284 285 286 287 288 289 290 291
        # 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
292 293
        }
      }
294 295
      $sth->finish();

296 297
      $cs = $high_cs;

298
      #return if we did not find any appropriate match:
299
      return undef if(!defined($high_ver));
300 301
    } else {

302 303
      my ($id, $cs_id);
      ($seq_region_name, $id, $length, $cs_id) = $sth->fetchrow_array();
304
      $sth->finish();
305

306 307
      # cahce to speed up for future queries
      my $arr = [ $id, $seq_region_name, $cs_id, $length ];
308 309
      $self->{'sr_name_cache'}->{"$seq_region_name:$cs_id"} = $arr;
      $self->{'sr_id_cache'}->{"$id"} = $arr;
310 311

      $cs = $csa->fetch_by_dbID( $cs_id );
312
    }
313
  }
314

315
  $end = $length if(!defined($end));
Graham McVicker's avatar
Graham McVicker committed
316

317 318
  if($end+1 < $start) {
    throw("start [$start] must be less than or equal to end+1 [$end+1]");
319 320
  }

321
  return Bio::EnsEMBL::Slice->new(-COORD_SYSTEM      => $cs,
322 323 324 325 326 327
                                  -SEQ_REGION_NAME   => $seq_region_name,
                                  -SEQ_REGION_LENGTH => $length,
                                  -START             => $start,
                                  -END               => $end,
                                  -STRAND            => $strand,
                                  -ADAPTOR           => $self);
328 329 330 331
}



332

333
=head2 fetch_by_name
334

Graham McVicker's avatar
Graham McVicker committed
335
  Arg [1]    : string $name
336 337 338 339 340 341 342 343 344
  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
345

346 347 348 349
               Returns undef if no seq_region with the provided name exists in
               the database.

  Returntype : Bio::EnsEMBL::Slice or undef
350 351
  Exceptions : throw if incorrent arg provided
  Caller     : Pipeline
352
  Status     : Stable
353 354 355

=cut

356 357 358
sub fetch_by_name {
  my $self = shift;
  my $name = shift;
Graham McVicker's avatar
Graham McVicker committed
359

360 361
  if(!$name) {
    throw("name argument is required");
Graham McVicker's avatar
Graham McVicker committed
362
  }
363 364 365

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

366
  if(scalar(@array) < 3 || scalar(@array) > 6) {
367
    throw("Malformed slice name [$name].  Format is " .
Laura Clarke's avatar
 
Laura Clarke committed
368
        "coord_system:version:name:start:end:strand");
369 370
  }

371
  # Rearrange arguments to suit fetch_by_region
372

373 374 375 376 377 378 379 380 381
  my @targetarray;

  $targetarray[0]=$array[0];
  $targetarray[5]=(($array[1]&&$array[1] ne "")?$array[1]:undef);
  $targetarray[1]=(($array[2]&&$array[2] ne "")?$array[2]:undef);
  $targetarray[2]=(($array[3]&&$array[3] ne "")?$array[3]:undef);
  $targetarray[3]=(($array[4]&&$array[4] ne "")?$array[4]:undef);
  $targetarray[4]=(($array[5]&&$array[5] ne "")?$array[5]:undef);
  return $self->fetch_by_region(@targetarray);
Graham McVicker's avatar
Graham McVicker committed
382
}
Graham McVicker's avatar
Graham McVicker committed
383 384 385



386 387 388 389 390 391 392 393
=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.
394 395
               Returns undef if no such slice exists.
  Returntype : Bio::EnsEMBL::Slice or undef
Graham McVicker's avatar
Graham McVicker committed
396 397
  Exceptions : none
  Caller     : general
398
  Status     : Stable
Graham McVicker's avatar
Graham McVicker committed
399 400 401

=cut

402 403 404
sub fetch_by_seq_region_id {
  my ($self, $seq_region_id) = @_;

405
  my $arr = $self->{'sr_id_cache'}->{ $seq_region_id };
406 407
  my ($name, $length, $cs);

408 409 410 411
  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);
412 413 414 415 416
  } else {
    my $sth = $self->prepare("SELECT name, length, coord_system_id " .
                             "FROM seq_region " .
                             "WHERE seq_region_id = ?");

417 418
    $sth->bind_param(1,$seq_region_id,SQL_INTEGER);
    $sth->execute();
419

420
    return undef if($sth->rows() == 0);
421 422 423 424 425 426 427 428

    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
429
    my $arr = [ $seq_region_id, $name, $cs_id, $length ];
430 431 432

    $self->{'sr_name_cache'}->{"$name:$cs_id"} = $arr;
    $self->{'sr_id_cache'}->{"$seq_region_id"} = $arr;
433 434
  }

435 436 437 438 439 440 441
  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
442 443
}

Graham McVicker's avatar
Graham McVicker committed
444

445 446

=head2 get_seq_region_id
Graham McVicker's avatar
Graham McVicker committed
447

Arne Stabenau's avatar
Arne Stabenau committed
448
  Arg [1]    : Bio::EnsEMBL::Slice $slice
449 450 451 452 453 454 455 456 457 458
               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
459
  Status     : Stable
Arne Stabenau's avatar
Arne Stabenau committed
460 461

=cut
Graham McVicker's avatar
Graham McVicker committed
462

463 464 465
sub get_seq_region_id {
  my $self = shift;
  my $slice = shift;
Graham McVicker's avatar
Graham McVicker committed
466

467 468 469
  if(!$slice || !ref($slice) || !$slice->isa('Bio::EnsEMBL::Slice')) {
    throw('Slice argument is required');
  }
470
  
471
  my $seq_region_name = $slice->seq_region_name();
472
  my $key = $seq_region_name.":".$slice->coord_system->dbID();
473
  my $arr = $self->{'sr_name_cache'}->{"$key"};
474

475 476
  if( $arr ) {
    return $arr->[0];
477 478
  }

479
  my $cs_id = $slice->coord_system->dbID();
480 481 482 483 484 485

  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
486 487 488
  $sth->bind_param(1,"$seq_region_name",SQL_VARCHAR);
  $sth->bind_param(2,$cs_id,SQL_INTEGER);
  $sth->execute();
489 490 491

  if($sth->rows() != 1) {
    throw("Non existant or ambigous seq_region:\n" .
492 493 494
          "  coord_system=[$cs_id],\n" .
          "  name=[$seq_region_name],\n");

495 496 497 498 499 500
  }

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

  #cache information for future requests
501
  $arr = [ $seq_region_id, $seq_region_name, $cs_id, $length ];
502 503 504 505

  $self->{'sr_name_cache'}->{"$seq_region_name:$cs_id"} = $arr;
  $self->{'sr_id_cache'}->{"$seq_region_id"} = $arr;

506
  return $seq_region_id;
Arne Stabenau's avatar
Arne Stabenau committed
507
}
508 509


Arne Stabenau's avatar
Arne Stabenau committed
510

511 512 513 514 515 516 517 518 519
=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
520
  Arg [3]    : bool $include_non_reference (optional)
521 522 523
               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
524 525
  Arg [4]    : int $include_duplicates (optional)
               If set duplicate regions will be returned.
526

527 528 529
  Example    : @chromos = @{$slice_adaptor->fetch_all('chromosome','NCBI33')};
               @contigs = @{$slice_adaptor->fetch_all('contig')};

Graham McVicker's avatar
Graham McVicker committed
530 531 532 533 534
               # 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)};
535 536 537 538 539 540

  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.
541 542
               If the coordinate system with the provided name and version
               does not exist an empty list is returned.
543
               If the coordinate system name provided is 'toplevel', all
544
               non-redundant toplevel slices are returned.
545 546 547 548

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

549
  Returntype : listref of Bio::EnsEMBL::Slices
550
  Exceptions : none
551
  Caller     : general
552
  Status     : Stable
Web Admin's avatar
Web Admin committed
553 554 555

=cut

556 557 558 559 560
sub fetch_all {
  my $self = shift;
  my $cs_name = shift;
  my $cs_version = shift || '';

561
  my ($include_non_reference, $include_duplicates) = @_;
Web Admin's avatar
Web Admin committed
562

563 564 565
  #
  # verify existance of requested coord system and get its id
  #
566 567 568 569 570 571
  my $csa       = $self->db->get_CoordSystemAdaptor();
  my $orig_cs   = $csa->fetch_by_name($cs_name, $cs_version);

  return [] if(!$orig_cs);

  my $sth;
572

573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591
  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;
    }
    $sth2->finish();
  }
592
  #
593
  # Retrieve the seq_regions from the database
594
  #
595
  if($orig_cs->is_top_level()) {
596 597 598 599 600 601 602 603
   $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");
604 605
    $sth->execute();
  } else {
606 607 608 609
     $sth =
       $self->prepare('SELECT seq_region_id, name, length, coord_system_id ' .
                      'FROM   seq_region ' .
                      'WHERE  coord_system_id =?');
610 611
     $sth->bind_param(1,$orig_cs->dbID,SQL_INTEGER);
     $sth->execute();
612
  }
613

614 615
  my ($seq_region_id, $name, $length, $cs_id);
  $sth->bind_columns(\$seq_region_id, \$name, \$length, \$cs_id);
616

617
  my $cache_count = 0;
618 619 620

  my @out;
  while($sth->fetch()) {
621 622
    if(!defined($bad_vals{$seq_region_id})){
      my $cs = $csa->fetch_by_dbID($cs_id);
623

624
      if(!$cs) {
625
        throw("seq_region $name references non-existent coord_system $cs_id.");
626
      }
627

628 629
      #cache values for future reference, but stop adding to the cache once we
      #we know we have filled it up
630
      if($cache_count < $Bio::EnsEMBL::Utils::SeqRegionCache::SEQ_REGION_CACHE_SIZE) {
631
        my $arr = [ $seq_region_id, $name, $cs_id, $length ];
632 633 634

        $self->{'sr_name_cache'}->{"$name:$cs_id"} = $arr;
        $self->{'sr_id_cache'}->{"$seq_region_id"} = $arr;
635

636
        $cache_count++;
637
      }
638 639 640 641 642 643 644 645 646 647 648

      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){
649 650 651 652 653 654 655 656 657 658 659 660
        # 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];
            }
661
          }
662 663 664
        } else {
          # no duplicate regions
          push @out, $slice;
665
        }
666 667
      } else {
        # we want duplicates anyway so do not do any checks
668
        push @out, $slice;
669
      }
670
    }
671
  }
672
  $sth->finish();
673

674
  return \@out;
Web Admin's avatar
Web Admin committed
675 676 677
}


678

Web Admin's avatar
Web Admin committed
679 680 681 682 683 684
=head2 fetch_by_band

 Title   : fetch_by_band
 Usage   :
 Function: create a Slice representing a series of bands
 Example :
685
 Returns : Bio::EnsEMBL::Slice
Web Admin's avatar
Web Admin committed
686
 Args    : the band name
687
 Status     : Stable
Web Admin's avatar
Web Admin committed
688 689 690 691 692 693 694 695 696 697 698

=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");

699 700
  $sth->bind_param(1,"$band%",SQL_VARCHAR);
  $sth->execute();
Web Admin's avatar
Web Admin committed
701 702 703 704 705 706 707 708 709 710
  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");
}

711 712 713 714 715 716
=head2 fetch_by_chr_band

 Title   : fetch_by_chr_band
 Usage   :
 Function: create a Slice representing a series of bands
 Example :
717
 Returns : Bio::EnsEMBL::Slice
718
 Args    : the band name
719
 Status     : Stable
720 721 722 723 724 725

=cut

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

Web Admin's avatar
Web Admin committed
726
  my $chr_slice = $self->fetch_by_region('toplevel', $chr);
727 728 729
  my $seq_region_id = $self->get_seq_region_id($chr_slice);

  my $sth = $self->db->prepare
Web Admin's avatar
Web Admin committed
730
        ("select min(k.seq_region_start), max(k.seq_region_end) " .
731 732 733
         "from karyotype as k " .
         "where k.seq_region_id = ? and k.band like ?");

734 735 736
  $sth->bind_param(1,$seq_region_id,SQL_INTEGER);
  $sth->bind_param(2,"$band%",SQL_VARCHAR);
  $sth->execute();
737 738 739
  my ( $slice_start, $slice_end) = $sth->fetchrow_array;

  if(defined $slice_start) {
Web Admin's avatar
Web Admin committed
740
    return $self->fetch_by_region('toplevel',$chr,$slice_start,$slice_end);
741 742 743
  }

  throw("Band not recognised in database");
744 745 746 747
}



Web Admin's avatar
Web Admin committed
748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
=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
764
  Status     : Stable
Web Admin's avatar
Web Admin committed
765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780

=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
781
=head2 fetch_by_transcript_stable_id
782

Graham McVicker's avatar
Graham McVicker committed
783 784 785 786 787 788 789 790 791
  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 
792 793
               number of basepairs on either side of the
               transcript.
Graham McVicker's avatar
Graham McVicker committed
794
  Returntype : Bio::EnsEMBL::Slice
795
  Exceptions : Thrown if the transcript is not in the database.
Graham McVicker's avatar
Graham McVicker committed
796
  Caller     : general
797
  Status     : Stable
798 799 800

=cut

Graham McVicker's avatar
Graham McVicker committed
801
sub fetch_by_transcript_stable_id{
802 803
  my ($self,$transcriptid,$size) = @_;

804 805
  throw('Transcript argument is required.') if(!$transcriptid);

806
  my $ta = $self->db->get_TranscriptAdaptor;
807
  my $transcript = $ta->fetch_by_stable_id($transcriptid);
808

809 810 811 812
  throw("Transcript [$transcriptid] does not exist in DB.") if(!$transcript);

  return $self->fetch_by_Feature($transcript, $size);
}
813

Graham McVicker's avatar
Graham McVicker committed
814 815


Graham McVicker's avatar
Graham McVicker committed
816 817
=head2 fetch_by_transcript_id

Graham McVicker's avatar
Graham McVicker committed
818 819 820 821 822 823 824
  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);
825 826 827 828
  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
829
  Returntype : Bio::EnsEMBL::Slice
830 831
  Exceptions : throw on incorrect args
               throw if transcript is not in database
Graham McVicker's avatar
Graham McVicker committed
832
  Caller     : general
833
  Status     : Stable
834 835 836

=cut

Graham McVicker's avatar
Graham McVicker committed
837
sub fetch_by_transcript_id {
838
  my ($self,$transcriptid,$size) = @_;
Graham McVicker's avatar
Graham McVicker committed
839

840
  throw('Transcript id argument is required.') if(!$transcriptid);
Graham McVicker's avatar
Graham McVicker committed
841

842 843 844 845 846 847
  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);
848 849
}

850 851


Graham McVicker's avatar
Graham McVicker committed
852
=head2 fetch_by_gene_stable_id
853

Graham McVicker's avatar
Graham McVicker committed
854
  Arg [1]    : string $geneid
855
               The stable id of the gene around which the slice is
Graham McVicker's avatar
Graham McVicker committed
856 857
               desired
  Arg [2]    : (optional) int $size
Graham McVicker's avatar
Graham McVicker committed
858
               The length of the flanking regions the slice should encompass
Graham McVicker's avatar
Graham McVicker committed
859 860
               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
861 862 863
  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.
864
               The slice will be created in the genes native coordinate system.
Graham McVicker's avatar
Graham McVicker committed
865
  Returntype : Bio::EnsEMBL::Slice
866 867
  Exceptions : throw on incorrect args
               throw if transcript does not exist
Graham McVicker's avatar
Graham McVicker committed
868
  Caller     : general
869
  Status     : Stable
870 871 872

=cut

873 874
sub fetch_by_gene_stable_id {
  my ($self,$geneid,$size) = @_;
875

876
  throw('Gene argument is required.') if(!$geneid);
877

878 879
  my $gene_adaptor = $self->db->get_GeneAdaptor();
  my $gene = $gene_adaptor->fetch_by_stable_id($geneid);
880

881
  throw("Gene [$geneid] does not exist in DB.") if(!$gene);
882

883
  return $self->fetch_by_Feature($gene, $size);
884 885 886
}


887

888
=head2 fetch_by_Feature
Graham McVicker's avatar
Graham McVicker committed
889

890 891 892 893
  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.
894 895
               The size may also be provided as a percentage of the feature 
               size such as 200% or 80.5%.
896 897 898 899 900 901
  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
902
  Returntype : Bio::EnsEMBL::Slice
903 904 905 906
  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
907
  Status     : Stable
Graham McVicker's avatar
Graham McVicker committed
908 909 910

=cut

911 912
sub fetch_by_Feature{
  my ($self, $feature, $size) = @_;
Graham McVicker's avatar
Graham McVicker committed
913

914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945
  $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) {