tripal/tripal_chado/includes/TripalFields/data__sequence/data__sequence.inc

<?php

class data__sequence extends ChadoField {


  // --------------------------------------------------------------------------
  //                     EDITABLE STATIC CONSTANTS
  //
  // The following constants SHOULD be set for each descendent class.  They are
  // used by the static functions to provide information to Drupal about
  // the field and it's default widget and formatter.
  // --------------------------------------------------------------------------

  // The default lable for this field.
  public static $default_label = 'Sequence';

  // The default description for this field.
  public static $description = 'A field for managing nucleotide and protein residues.';

  // Provide a list of instance specific settings. These can be access within
  // the instanceSettingsForm.  When the instanceSettingsForm is submitted
  // then Drupal with automatically change these settings for the instnace.
  // It is recommended to put settings at the instance level whenever possible.
  // If you override this variable in a child class be sure to replicate the
  // term_name, term_vocab, term_accession and term_fixed keys as these are
  // required for all TripalFields.
  public static $default_instance_settings  = array(
    // The short name for the vocabulary (e.g. shcema, SO, GO, PATO, etc.).
    'term_vocabulary' => 'data',
    // The name of the term.
    'term_name' => 'sequence',
    // The unique ID (i.e. accession) of the term.
    'term_accession' => '2044',
    // Set to TRUE if the site admin is allowed to change the term
    // type. This will create form elements when editing the field instance
    // to allow the site admin to change the term settings above.
    'term_fixed' => FALSE,
  );

  // The default widget for this field.
  public static $default_widget = 'data__sequence_widget';

  // The default formatter for this field.
  public static $default_formatter = 'data__sequence_formatter';

  /**
   * @see TripalField::load()
   */
  public function load($entity, $details = array()) {
    $field_name = $this->field['field_name'];
    $feature = $details['record'];

    // We don't want to get the sequence for traditionally large types. They are
    // too big,  bog down the web browser, take longer to load and it's not
    // reasonable to print them on a page.
    if(strcmp($feature->type_id->name,'scaffold') != 0 and
       strcmp($feature->type_id->name,'chromosome') != 0 and
       strcmp($feature->type_id->name,'supercontig') != 0 and
       strcmp($feature->type_id->name,'pseudomolecule') != 0) {
      $feature = chado_expand_var($feature,'field','feature.residues');
      $entity->{$field_name}['und'][0]['value'] = $feature->residues;
    }

 /*    // Add in sequences from alignments.
    $options = array(
      'return_array' => 1,
      'include_fk' => array(
        'srcfeature_id' => array(
          'type_id' => 1
        ),
        'feature_id' => array(
          'type_id' => 1
        ),
      ),
    );
    $feature = chado_expand_var($feature, 'table', 'featureloc', $options);
    $featureloc_sequences = $this->get_featureloc_sequences($feature->feature_id, $feature->featureloc->feature_id);

    // Add in the coding sequences. It's faster to provide the SQL rather than
    // to use chado_generate_var based on the type.
    $sql = "
      SELECT F.*
      FROM {feature_relationship} FR
        INNER JOIN {feature} F on FR.subject_id = F.feature_id
        INNER JOIN {cvterm} CVT on CVT.cvterm_id = F.type_id
        INNER JOIN {cvterm} RCVT on RCVT.cvterm_id = FR.type_id
        INNER JOIN {featureloc} FL on FL.feature_id = F.feature_id
      WHERE
        FR.object_id = :feature_id and
        CVT.name = 'CDS' and
        RCVT.name = 'part_of'
      ORDER BY FR.rank ASC
    ";
    $results = chado_query($sql, array(':feature_id' => $feature->feature_id));
    $coding_seq = '';
    while ($CDS = $results->fetchObject()) {
      if ($CDS->residues) {
        $coding_seq .= $CDS->residues;
      }
    }
    if ($coding_seq) {
      $entity->{$field_name}['und'][$num_seqs++]['value'] = array(
        '@type' => 'SO:0000316',
        'type' => 'coding_sequence',
        'label' => 'Coding sequence (CDS)',
        'defline' => tripal_get_fasta_defline($feature, 'CDS', NULL, '', strlen($coding_seq)),
        'residues' => $coding_seq,
      );
    }

    foreach($featureloc_sequences as $src => $attrs){
      // the $attrs array has the following keys
      //   * id:  a unique identifier combining the feature id with the cvterm id
      //   * type: the type of sequence (e.g. mRNA, etc)
      //   * location:  the alignment location
      //   * defline: the definition line
      //   * formatted_seq: the formatted sequences
      //   * featureloc:  the feature object aligned to
      $entity->{$field_name}['und'][$num_seqs++]['value'] = array(
        'residues' => $attrs['residues'],
        '@type' => 'SO:0000110',
        'type' => 'sequence_feature',
        'defline' => tripal_get_fasta_defline($feature, '', $attrs['featureloc'], 'CDS', strlen($attrs['residues'])),
        'label' => 'Sequence from alignment at ' . $attrs['location'],
      );


      // check to see if this alignment has any CDS. If so, generate a CDS sequence
      $cds_sequence = tripal_get_feature_sequences(
          array(
            'feature_id' => $feature->feature_id,
            'parent_id' => $attrs['featureloc']->srcfeature_id->feature_id,
            'name' => $feature->name,
            'featureloc_id' => $attrs['featureloc']->featureloc_id,
          ),
          array(
            'derive_from_parent' => 1, // CDS are in parent-child relationships so we want to use the sequence from the parent
            'aggregate' => 1, // we want to combine all CDS for this feature into a single sequence
            'sub_feature_types' => array('CDS'), // we're looking for CDS features
            'is_html' => 0
          )
          );

      if (count($cds_sequence) > 0) {
        // the tripal_get_feature_sequences() function can return multiple sequences
        // if a feature is aligned to multiple places. In the case of CDSs we expect
        // that one mRNA is only aligned to a single location on the assembly so we
        // can access the CDS sequence with index 0.
        if ($cds_sequence[0]['residues']) {
          $entity->{$field_name}['und'][$num_seqs++]['value'] = array(
            'residues' => $cds_sequence[0]['residues'],
            '@type' => 'SO:0000316',
            'type' => 'coding_sequence',
            'defline' => tripal_get_fasta_defline($feature, '', $attrs['featureloc'], 'CDS', $cds_sequence[0]['length']),
            'label' => 'Coding sequence (CDS) from alignment at  ' . $attrs['location'],
          );
        }
      }
    } */
  }
}