Skip to content

Commit

Permalink
Merge pull request #1136 from tripal/1135-tv3-sequence_fields
Browse files Browse the repository at this point in the history 
Fix for bugs with displaying sequences
  • Loading branch information
@laceysanderson
laceysanderson committed Dec 20, 2020
2 parents 48f098f + 51cdd14 commit 861164d
Show file tree
Hide file tree
Showing 16 changed files with 678 additions and 177 deletions.
12 changes: 7 additions & 5 deletions tripal/includes/tripal.fields.inc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -911,11 +911,13 @@ function tripal_form_field_ui_display_overview_form_alter(&$form, &$form_state,
$fields_names = element_children($form['fields']);
foreach ($fields_names as $field_name) {
$field_info = field_info_field($field_name);
if ($field_info['type'] == 'kvproperty_adder') {
unset($form['fields'][$field_name]);
}
if ($field_info['type'] == 'cvterm_class_adder') {
unset($form['fields'][$field_name]);
if ($field_info) {
if ($field_info['type'] == 'kvproperty_adder') {
unset($form['fields'][$field_name]);
}
if ($field_info['type'] == 'cvterm_class_adder') {
unset($form['fields'][$field_name]);
}
}
}
}
Expand Down
8 changes: 1 addition & 7 deletions tripal_chado/includes/TripalFields/data__protein_sequence/data__protein_sequence.inc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -93,18 +93,12 @@ class data__protein_sequence extends ChadoField {
WHERE
FR.object_id = :feature_id and
CVT.name = 'polypeptide' and
RCVT.name IN ('derives_from', 'part_of')
RCVT.name IN ('derives_from', 'part_of')
ORDER BY FR.rank ASC
";
$proteins = chado_query($sql, [':feature_id' => $feature->feature_id]);
while ($protein = $proteins->fetchObject()) {
$entity->{$field_name}['und'][$num_seqs]['value'] = $protein->residues;
// Because we'll be saving a feature we need to maintain all of it's
// columns in the feature table. The following will add them all.
$columns = get_object_vars($protein);
foreach ($columns as $colname => $value) {
$entity->{$field_name}['und'][$num_seqs]['chado-feature__' . $colname] = $value;
}
$num_seqs++;
}
}
Expand Down
3 changes: 2 additions & 1 deletion ...l_chado/includes/TripalFields/data__protein_sequence/data__protein_sequence_formatter.inc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,8 @@ class data__protein_sequence_formatter extends ChadoFieldFormatter {
* @see TripalFieldFormatter::view()
*/
public function view(&$element, $entity_type, $entity, $langcode, $items, $display) {
$content = 'There is no protein sequence.';

$content = 'There is no protein sequence available.';
if (count($items) > 0 and $items[0]['value']) {
$num_bases = 50;
$content = '<pre class="protein-residues-formatter">';
Expand Down
99 changes: 1 addition & 98 deletions tripal_chado/includes/TripalFields/data__sequence/data__sequence.inc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ class data__sequence extends ChadoField {
public static $default_label = 'Sequence';

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

// Provide a list of instance specific settings. These can be accessed within
// the instanceSettingsForm. When the instanceSettingsForm is submitted
Expand Down Expand Up @@ -77,103 +77,6 @@ class data__sequence extends ChadoField {

$feature = chado_expand_var($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'][$num_seqs++]['value'] = array(
'@type' => 'SO:0000316',
'type' => 'coding_sequence',
'label' => 'Coding sequence (CDS)',
'defline' => chado_get_fasta_defline($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' => $attrs['residues'],
'@type' => 'SO:0000110',
'type' => 'sequence_feature',
'defline' => chado_get_fasta_defline($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 = chado_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 chado_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' => $cds_sequence[0]['residues'],
'@type' => 'SO:0000316',
'type' => 'coding_sequence',
'defline' => chado_get_fasta_defline($feature, '', $attrs['featureloc'], 'CDS', $cds_sequence[0]['length']),
'label' => 'Coding sequence (CDS) from alignment at ' . $attrs['location'],
);
}
}
} */
}
}

76 changes: 40 additions & 36 deletions tripal_chado/includes/TripalFields/data__sequence_coordinates/data__sequence_coordinates.inc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -224,49 +224,53 @@ class data__sequence_coordinates extends ChadoField {
$strand_term = chado_get_semweb_term('featureloc', 'strand');
$phase_term = chado_get_semweb_term('featureloc', 'phase');

$options = [
'return_array' => TRUE,
'order_by' => ['rank' => 'ASC'],
];
$feature = chado_expand_var($feature, 'table', 'featureloc', $options);

// Set some defauls for the empty record
$entity->{$field_name}['und'][0] = [
'value' => '',
];

// Get the featureloc records that this feature is aligned to.
$aligned = $feature->featureloc->feature_id;
if ($aligned) {
foreach ($aligned as $index => $featureloc) {
$srcfeature = $featureloc->srcfeature_id->name;
// Get the featureloc records that this feature is aligned to. We use
// this SQL rather than the chado_expand_var function because we don't
// want the residues included from the srcfeature_id which may be huge
// and overrun memory.
$featurelocs_sql = "
SELECT SRCF.name, FL.srcfeature_id, FL.strand, FL.fmin, FL.fmax, FL,phase
FROM {featureloc} FL
INNER JOIN {feature} SRCF on SRCF.feature_id = FL.srcfeature_id
WHERE FL.feature_id = :feature_id
ORDER BY rank ASC
";
$aligned = chado_query($featurelocs_sql, [':feature_id' => $feature->feature_id]);
$index = 0;
while ($featureloc = $aligned->fetchObject()) {
$srcfeature = $featureloc->name;
$strand = '';
if ($featureloc->strand == 1) {
$strand = '+';
}
elseif ($featureloc->strand == -1) {
$strand = '-';
}
else {
$strand = '';
if ($featureloc->strand == 1) {
$strand = '+';
}
elseif ($featureloc->strand == -1) {
$strand = '-';
}
else {
$strand = '';
}
$fmin = $featureloc->fmin + 1;
$fmax = $featureloc->fmax;
$entity->{$field_name}['und'][$index] = [
'value' => [
$description => $srcfeature . ':' . $fmin . '-' . $fmax . $strand,
$reference_term => $srcfeature,
$fmin_term => $fmin,
$fmax_term => $fmax,
$strand_term => $strand,
$phase_term => $featureloc->phase,
],
];
$sentity_id = chado_get_record_entity_by_table('feature_id', $featureloc->srcfeature_id->feature_id);
if ($sentity_id) {
$entity->{$field_name}['und'][0]['value']['entity'] = 'TripalEntity:' . $sentity_id;
}
}
$fmin = $featureloc->fmin + 1;
$fmax = $featureloc->fmax;
$entity->{$field_name}['und'][$index] = [
'value' => [
$description => $srcfeature . ':' . $fmin . '-' . $fmax . $strand,
$reference_term => $srcfeature,
$fmin_term => $fmin,
$fmax_term => $fmax,
$strand_term => $strand,
$phase_term => $featureloc->phase,
],
];
$sentity_id = chado_get_record_entity_by_table('feature_id', $featureloc->srcfeature_id);
if ($sentity_id) {
$entity->{$field_name}['und'][0]['value']['entity'] = 'TripalEntity:' . $sentity_id;
}
$index++;
}
}
}

0 comments on commit 861164d

Please sign in to comment.