$analysis_id, 'feature_id' => $feature_id);
$result = tripal_core_chado_select('analysisfeature', $columns, $values);
$analysisfeature_id = $result[0]->analysisfeature_id;
}
// get the property.
return tripal_core_get_property('analysisfeature', $analysisfeature_id, $property, $cv_name);
}
/**
* Insert a property for an analysis feature
*
* @param $analysis_id
* The analysis ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with a feature ID.
* @param $feature_id
* The feature ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with an analysis ID.
* @param $analysisfeature_id
* The analysis feature ID for the analysis feature. This argument is
* optional and can be used rather than specifying the $analysis_id and
* $feature_id arguments. If all three arguments are specified (e.g.
* an $analysis_id, $feature_id and $analysisfeature_id, then the
* $analysisfeature_id is used and the other two arguments are ignored.
* @param $property
* The cvterm name of the property to insert
* @param $value
* The value of the property to insert
* @param $update_if_present
* A boolean indicated whether to update the record if it's already present
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_analysis_insert_property($analysis_id = NULL, $feature_id = NUll,
$analysisfeature_id = NULL, $property, $value, $update_if_present = 0, $cv_name = 'tripal') {
// check that the incoming arguments are correct
if (($analysis_id and !$feature_id) or
(!$analysis_id and $feature_id)) {
watchdog('tripal_feature',
'tripal_feature_analysis_insert_property: Both an analysis ID and feature ID should be specified',
array(), WATCHDOG_WARNING);
}
// get the analysisfeature_id if one is not provided
if (!$analysisfeature_id) {
$columns = array('analysisfeature_id');
$values = array('analysis_id' => $analysis_id, 'feature_id' => $feature_id);
$result = tripal_core_chado_select('analysisfeature', $columns, $values);
$analysisfeature_id = $result[0]->analysisfeature_id;
}
// insert the property.
$success = tripal_core_insert_property('analysisfeature', $analysisfeature_id,
$property, $cv_name, $value, $update_if_present);
if (!$success) {
watchdog('tripal_feature',
'tripal_feature_analysis_insert_property: Failed to insert analysis feature property',
array(), WATCHDOG_WARNING);
return FALSE;
}
return $success;
}
/**
* Update an analysis feature property using the property name. Use this
* when a property only exists once for a given analysis feature. When more
* than one value can exist for the same property use the
* tripal_feature_analysis_update_property_by_id() function.
*
* @param $analysis_id
* The analysis ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with a feature ID.
* @param $feature_id
* The feature ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with an analysis ID.
* @param $analysisfeature_id
* The analysis feature ID for the analysis feature. This argument is
* optional and can be used rather than specifying the $analysis_id and
* $feature_id arguments. If all three arguments are specified (e.g.
* an $analysis_id, $feature_id and $analysisfeature_id, then the
* $analysisfeature_id is used and the other two arguments are ignored.
* @param $property
* The cvterm name of the property to update
* @param $value
* The value of the property to update
* @param $insert_if_missing
* A boolean indicated whether to insert the record if it's absent
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* Note: The property will be identified using the unique combination of the $analysis_id and $property
* and then it will be updated with the supplied value
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_analysis_update_property($analysis_id = NULL, $feature_id = NUll,
$analysisfeature_id = NULL, $property, $value, $insert_if_missing = 0,
$cv_name = 'tripal') {
// check that the incoming arguments are correct
if (($analysis_id and !$feature_id) or
(!$analysis_id and $feature_id)) {
watchdog('tripal_feature',
'tripal_feature_analysis_update_property: Both an analysis ID and feature ID should be specified',
array(), WATCHDOG_WARNING);
}
// get the analysisfeature_id if one is not provided
if (!$analysisfeature_id) {
$columns = array('analysisfeature_id');
$values = array('analysis_id' => $analysis_id, 'feature_id' => $feature_id);
$result = tripal_core_chado_select('analysisfeature', $columns, $values);
$analysisfeature_id = $result[0]->analysisfeature_id;
}
// update the property.
return tripal_core_update_property('analysisfeature', $analysisfeature_id, $property, $cv_name, $value, $insert_if_missing);
}
/**
* Update a property for an analysis feature using the analysisfeatureprop_id.
*
* @param $analysisfeatureprop_id
* The analysis feature property ID for the analysis feature.
* @param $property
* The cvterm name of the property
* @param $value
* The value of the property
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
* *
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_analysis_update_property_by_id($analysisfeatureprop_id,
$property, $value, $cv_name = 'tripal') {
// update the property.
return tripal_core_update_property_by_id('analysisfeature',
$analysisfeatureprop_id, $property, $cv_name, $value);
}
/**
* Delete an analysis feature property using the property name. Use this
* when a property only exists once for a given analysis feature. When more
* than one value can exist for the same property use the
* tripal_feature_analysis_delete_property_by_id() function.
*
* @param $analysis_id
* The analysis ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with a feature ID.
* @param $feature_id
* The feature ID for the analysis feature. This argument is optional but
* if specified it must also be accompanied with an analysis ID.
* @param $analysisfeature_id
* The analysis feature ID for the analysis feature. This argument is
* optional and can be used rather than specifying the $analysis_id and
* $feature_id arguments. If all three arguments are specified (e.g.
* an $analysis_id, $feature_id and $analysisfeature_id, then the
* $analysisfeature_id is used and the other two arguments are ignored.
* @param $property
* The cvterm name of the property to delete
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* Note: The property will be identified using the unique combination of the $analysis_id and $property
* and then it will be deleted
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_analysis_delete_property($analysis_id = NULL, $feature_id = NUll,
$analysisfeature_id = NULL, $property, $cv_name = 'tripal') {
// check that the incoming arguments are correct
if (($analysis_id and !$feature_id) or
(!$analysis_id and $feature_id)) {
watchdog('tripal_feature',
'tripal_feature_analysis_delete_property: Both an analysis ID and feature ID should be specified',
array(), WATCHDOG_WARNING);
}
// get the analysisfeature_id if one is not provided
if (!$analysisfeature_id) {
$columns = array('analysisfeature_id');
$values = array('analysis_id' => $analysis_id, 'feature_id' => $feature_id);
$result = tripal_core_chado_select('analysisfeature', $columns, $values);
$analysisfeature_id = $result[0]->analysisfeature_id;
}
// get the property.
return tripal_core_delete_property('analysisfeature', $analysisfeature_id, $property, $cv_name);
}
/**
* Delete a property using the analysisfeatureprop_id
*
* @param $analysisfeatureprop_id
* The analysis feature property ID for the analysis feature.
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_analysis_delete_property_by_id($analysisfeatureprop_id) {
// get the property.
return tripal_core_delete_property_by_id('analysisfeature', $analysisfeatureprop_id);
}
/**
* Retrieve properties of a given type for a given feature
*
* @param $feature_id
* The feature_id of the properties you would like to retrieve
* @param $property
* The cvterm name of the properties to retrieve
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* @return
* A feature chado variable with the specified properties expanded
*
* @ingroup tripal_feature_api
*/
function tripal_feature_get_property($feature_id, $property, $cv_name='tripal') {
return tripal_core_get_property('feature', $feature_id, $property, $cv_name);
}
/**
* Insert a given property
*
* @param $feature_id
* The feature_id of the property to insert
* @param $property
* The cvterm name of the property to insert
* @param $value
* The value of the property to insert
* @param $update_if_present
* A boolean indicated whether to update the record if it's already present
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_insert_property($feature_id, $property, $value,
$update_if_present = 0, $cv_name = 'tripal') {
return tripal_core_insert_property('feature', $feature_id, $property,
$cv_name, $value, $update_if_present);
}
/**
* Update a feature property using the property name. Only use this
* if the property is unique and only exist once for the feature.
*
* @param $feature_id
* The feature_id of the property to update
* @param $property
* The cvterm name of the property to update
* @param $value
* The value of the property to update
* @param $insert_if_missing
* A boolean indicated whether to insert the record if it's absent
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* Note: The property will be identified using the unique combination of the $feature_id and $property
* and then it will be updated with the supplied value
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_update_property($feature_id, $property,
$value, $insert_if_missing = 0, $cv_name = 'tripal') {
return tripal_core_update_property('feature', $feature_id, $property, $cv_name, $value, $insert_if_missing);
}
/**
* Update a given feature property using the featureprop_id
*
* @param $featureprop_id
* The featureprop_id of the property to update
* @param $property
* The cvterm name of the property
* @param $value
* The value of the property
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_update_property_by_id($featureprop_id, $property,
$value, $cv_name = 'tripal') {
return tripal_core_update_property_by_id('feature', $featureprop_id, $property, $cv_name, $value);
}
/**
* Delete a given feature property using the property name. Only use this
* if the property is unique and only exists once for the feature.
*
* @param $feature_id
* The feature_id of the property to delete
* @param $property
* The cvterm name of the property to delete
* @param $cv_name
* Optional. The name of the cv to which the property belongs. By
* default this is the 'tripal' cv.
*
* Note: The property will be identified using the unique combination of the $feature_id and $property
* and then it will be deleted
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_delete_property($feature_id, $property, $cv_name='tripal') {
return tripal_core_delete_property('feature', $feature_id, $property, $cv_name);
}
/**
* Delete a given feature property using the featureprop_id
*
* @param $featureprop_id
* The feature_id of the property to delete
*
* @return
* True of success, False otherwise
*
* @ingroup tripal_feature_api
*/
function tripal_feature_delete_property_by_id($featureprop_id) {
return tripal_core_delete_property_by_id('feature', $featureprop_id);
}
/**
* Performs a reverse compliment of a nucleotide sequence
*
* @param $sequence
* The nucelotide sequence
*
* @return
* an upper-case reverse complemented sequence
*
* @ingroup tripal_feature_api
*/
function tripal_feature_reverse_complement($sequence) {
$seq = strtoupper($sequence);
$seq = strrev($seq);
$seq = str_replace("A", "t", $seq);
$seq = str_replace("T", "a", $seq);
$seq = str_replace("G", "c", $seq);
$seq = str_replace("C", "g", $seq);
$seq = str_replace("Y", "r", $seq);
$seq = str_replace("R", "y", $seq);
$seq = str_replace("W", "w", $seq);
$seq = str_replace("S", "s", $seq);
$seq = str_replace("K", "m", $seq);
$seq = str_replace("M", "k", $seq);
$seq = str_replace("D", "h", $seq);
$seq = str_replace("V", "b", $seq);
$seq = str_replace("H", "d", $seq);
$seq = str_replace("B", "v", $seq);
return strtoupper($seq);
}
/**
* Retrieves the sequence for a feature.
*
* @param $feature_id
* The feature_id of the feature for which the sequence will be retrieved
* @param $feature_name
* The feature name. This will appear on the FASTA definition line
* @param $num_bases_per_line
* Indicate the number of bases to use per line. A new line will be added
* after the specified number of bases on each line.
* @param $derive_from_parent
* Set to '1' if the sequence should be obtained from the parent to which
* this feature is aligned.
* @param $aggregate
* Set to '1' if the sequence should only contain sub features, excluding
* intra sub feature sequence. For example, set this option to obtain just
* the coding sequence of an mRNA.
* @param $output_format
* The type of format. Valid formats include 'fasta_html', 'fasta_txt' and
* 'raw'. The format 'fasta_txt' outputs line
* breaks as
tags and the entire return value is in a tag
* with a fixed-width font definition. 'fasta_txt' outputs line breaks with
* windows format carriage returns (e.g. \r\n) with no other formatting. The
* raw format is simply the sequence with now FASTA formatting and no
* line breaks.
* @param $upstream
* An integer specifing the number of upstream bases to include in the output
* @param $downstream
* An integer specifying the number of downstream bases to include in the
* output.
* @param $sub_features
* Only include sub features (or child features) of the types provided in the array
* @param $relationship
* If a relationship name is provided (e.g. sequence_of) then any sequences that
* are in relationships of this type with matched sequences are also included
* @param $rel_part
* If a relationship is provided in the preceeding argument then the rel_part
* must be either 'object' or 'subject' to indicate which side of the
* relationship the matched features belong
*
* @return
* The DNA/protein sequence formated as requested.
*
* @ingroup tripal_feature_api
*/
function tripal_feature_get_formatted_sequence($feature_id, $feature_name,
$num_bases_per_line, $derive_from_parent, $aggregate, $output_format,
$upstream, $downstream, $sub_features = array(), $relationship = '', $rel_part = '') {
// to speed things up we need to make sure we have a persistent connection
$connection = tripal_db_persistent_chado();
if (!$upstream) {
$upstream = 0;
}
if (!$downstream) {
$downstream = 0;
}
if ($rel_part == "object" or $rel_part == "subject") {
if($rel_part == "subject") {
$psql = '
PREPARE feature_rel_get_object (int, text) AS
SELECT FO.feature_id, FO.name, FO.uniquename, CVTO.name as feature_type, O.genus, O.species
FROM feature FS
INNER JOIN feature_relationship FR ON FR.subject_id = FS.feature_id
INNER JOIN cvterm CVTFR ON CVTFR.cvterm_id = FR.type_id
INNER JOIN feature FO ON FO.feature_id = FR.object_id
INNER JOIN cvterm CVTO ON CVTO.cvterm_id = FO.type_id
INNER JOIN organism O ON O.organism_id = FO.organism_id
WHERE
FS.feature_id = $1 AND
CVTFR.name = $2
';
$status = tripal_core_chado_prepare('feature_rel_get_object', $psql, array('int', 'text'));
if (!$status) {
watchdog('tripal_feature', "init: not able to prepare SQL statement '%name'",
array('%name' => 'feature_by_subject'), 'WATCHDOG ERROR');
}
$sql = "EXECUTE feature_rel_get_object(%d,'%s')";
$features = chado_query($sql, $feature_id, $relationship);
}
if($rel_part == "object") {
$psql = '
PREPARE feature_rel_get_subject (int, text) AS
SELECT FS.feature_id, FS.name, FS.uniquename, CVTO.name as feature_type, O.genus, O.species
FROM feature FO
INNER JOIN feature_relationship FR ON FR.object_id = FO.feature_id
INNER JOIN cvterm CVTFR ON CVTFR.cvterm_id = FR.type_id
INNER JOIN feature FS ON FS.feature_id = FR.subject_id
INNER JOIN cvterm CVTO ON CVTO.cvterm_id = FS.type_id
INNER JOIN organism O ON O.organism_id = FS.organism_id
WHERE
FO.feature_id = $1 AND
CVTFR.name = $2
';
$status = tripal_core_chado_prepare('feature_rel_get_subject', $psql, array('int', 'text'));
if (!$status) {
watchdog('tripal_feature', "init: not able to prepare SQL statement '%name'",
array('%name' => 'feature_by_object'), 'WATCHDOG ERROR');
}
$sql = "EXECUTE feature_rel_get_subject(%d,'%s')";
$features = chado_query($sql, $feature_id, $relationship);
}
$sequences = '';
while ($feature = db_fetch_object($features)) {
// recurse and get the sequences for these in the relationship
if ($rel_part == "subject") {
$defline = "$feature_name, $relationship, $feature->uniquename $feature->feature_type ($feature->genus $feature->species)";
}
if ($rel_part == "object") {
$defline = "$feature->uniquename $feature->feature_type ($feature->genus $feature->species), $relationship, $feature_name";
}
$sequences .= tripal_feature_get_formatted_sequence($feature->feature_id, $defline,
$num_bases_per_line, $derive_from_parent, $aggregate, $output_format,
$upstream, $downstream, $sub_features,'','');
}
return $sequences;
}
// prepare statements we'll need to use later
if (!tripal_core_is_sql_prepared('sequence_by_parent')) {
// prepare the queries we're going to use later during the render phase
// This SQL statement uses conditionals in the select clause to handle
// cases cases where the alignment is in the reverse direction and when
// the upstream and downstream extensions go beyond the lenght of the
// parent sequence.
$psql ='
PREPARE sequence_by_parent (int, int, int) AS
SELECT srcname, srcfeature_id, strand, srctypename, typename,
fmin, fmax, upstream, downstream, adjfmin, adjfmax,
substring(residues from (adjfmin + 1) for (upstream + (fmax - fmin) + downstream)) as residues,
genus, species
FROM (
SELECT
OF.name srcname, FL.srcfeature_id, FL.strand,
OCVT.name as srctypename, SCVT.name as typename,
FL.fmin, FL.fmax, OO.genus, OO.species,
CASE
WHEN FL.strand >= 0 THEN
CASE
WHEN FL.fmin - $1 <= 0 THEN 0
ELSE FL.fmin - $1
END
WHEN FL.strand < 0 THEN
CASE
WHEN FL.fmin - $2 <= 0 THEN 0
ELSE FL.fmin - $2
END
END as adjfmin,
CASE
WHEN FL.strand >= 0 THEN
CASE
WHEN FL.fmax + $2 > OF.seqlen THEN OF.seqlen
ELSE FL.fmax + $2
END
WHEN FL.strand < 0 THEN
CASE
WHEN FL.fmax + $1 > OF.seqlen THEN OF.seqlen
ELSE FL.fmax + $1
END
END as adjfmax,
CASE
WHEN FL.strand >= 0 THEN
CASE
WHEN FL.fmin - $1 <= 0 THEN FL.fmin
ELSE $1
END
ELSE
CASE
WHEN FL.fmax + $1 > OF.seqlen THEN OF.seqlen - FL.fmax
ELSE $1
END
END as upstream,
CASE
WHEN FL.strand >= 0 THEN
CASE
WHEN FL.fmax + $2 > OF.seqlen THEN OF.seqlen - FL.fmax
ELSE $2
END
ELSE
CASE
WHEN FL.fmin - $2 <= 0 THEN FL.fmin
ELSE $2
END
END as downstream,
OF.residues
FROM {featureloc} FL
INNER JOIN {feature} SF on FL.feature_id = SF.feature_id
INNER JOIN {cvterm} SCVT on SF.type_id = SCVT.cvterm_id
INNER JOIN {feature} OF on FL.srcfeature_id = OF.feature_id
INNER JOIN {cvterm} OCVT on OF.type_id = OCVT.cvterm_id
INNER JOIN {organism} OO on OF.organism_id = OO.organism_id
WHERE SF.feature_id = $3 and NOT (OF.residues = \'\' or OF.residues IS NULL)) as tbl1
';
$status = tripal_core_chado_prepare('sequence_by_parent', $psql, array('int', 'int', 'int'));
if (!$status) {
watchdog('tripal_feature',
"init: not able to prepare SQL statement '%name'",
array('%name' => 'sequence_by_parent'), 'WATCHDOG ERROR');
}
// this query is meant to get all of the sub features of any given
// feature (arg #1) and order them as they appear on the reference
// feature (arg #2).
$psql ='PREPARE sub_features (int, int) AS
SELECT SF.feature_id, CVT.name as type_name, SF.type_id
FROM {feature_relationship} FR
INNER JOIN {feature} SF on SF.feature_id = FR.subject_id
INNER JOIN {cvterm} CVT on CVT.cvterm_id = SF.type_id
INNER JOIN {featureloc} FL on FL.feature_id = FR.subject_id
INNER JOIN {feature} PF on PF.feature_id = FL.srcfeature_id
WHERE FR.object_id = $1 and PF.feature_id = $2
ORDER BY FL.fmin ASC';
$status = tripal_core_chado_prepare('sub_features', $psql, array('int', 'int'));
if (!$status) {
watchdog('tripal_views_handler_field_sequence',
"init: not able to prepare SQL statement '%name'",
array('%name' => 'ssub_features'), 'WATCHDOG ERROR');
}
$psql ='PREPARE count_sub_features (int, int) AS
SELECT count(*) as num_children
FROM {feature_relationship} FR
INNER JOIN {feature} SF on SF.feature_id = FR.subject_id
INNER JOIN {cvterm} CVT on CVT.cvterm_id = SF.type_id
INNER JOIN {featureloc} FL on FL.feature_id = FR.subject_id
INNER JOIN {feature} PF on PF.feature_id = FL.srcfeature_id
WHERE FR.object_id = $1 and PF.feature_id = $2';
$status = tripal_core_chado_prepare('count_sub_features', $psql, array('int', 'int'));
if (!$status) {
watchdog('tripal_views_handler_field_sequence',
"init: not able to prepare SQL statement '%name'",
array('%name' => 'count_sub_features'), 'WATCHDOG ERROR');
}
}
// if we need to get the sequence from the parent then do so now.
if ($derive_from_parent) {
// execute the query to get the sequence from the parent
$sql = "EXECUTE sequence_by_parent (%d, %d, %d)";
$parents = chado_query($sql, $upstream, $downstream, $feature_id);
while ($parent = db_fetch_object($parents)) {
$seq = ''; // initialize the sequence for each parent
// if we are to aggregate then we will ignore the feature returned
// by the query above and rebuild it using the sub features
if ($aggregate) {
// now get the sub features that are located on the parent.
$sql = "EXECUTE sub_features (%d, %d)";
$children = chado_query($sql, $feature_id, $parent->srcfeature_id);
$sql = "EXECUTE count_sub_features (%d, %d)";
$num_children = db_fetch_object(chado_query($sql, $feature_id, $parent->srcfeature_id));
// iterate through the sub features and concat their sequences. They
// should already be in order.
$types = array();
$i = 0;
while ($child = db_fetch_object($children)) {
// if the callee has specified that only certain sub features should be
// included then continue if this child is not one of those allowed
// subfeatures
if (count($sub_features) > 0 and !in_array($child->type_name, $sub_features)) {
$i++;
continue;
}
// keep up with the types
if (!in_array($child->type_name, $types)) {
$types[] = $child->type_name;
}
$sql = "EXECUTE sequence_by_parent (%d, %d, %d)";
// if the first sub feature we need to include the upstream bases. first check if
// the feature is in the foward direction or the reverse.
if ($i == 0 and $parent->strand >= 0) { // forward direction
// -------------------------- ref
// ....----> ---->
// up 1 2
$q = chado_query($sql, $upstream, 0, $child->feature_id);
}
elseif ($i == 0 and $parent->strand < 0) { // reverse direction
// -------------------------- ref
// ....<---- <----
// down 1 2
$q = chado_query($sql, 0, $downstream, $child->feature_id);
}
// Next, if the last sub feature we need to include the downstream bases. first check if
// the feature is in teh forward direction or the reverse
if ($i == $num_children->num_children - 1 and $parent->strand >= 0) { // forward direction
// -------------------------- ref
// ----> ---->....
// 1 2 down
$q = chado_query($sql, 0, $downstream, $child->feature_id);
}
elseif ($i == $num_children->num_children - 1 and $parent->strand < 0) { // reverse direction
// -------------------------- ref
// <---- <----....
// 1 2 up
$q = chado_query($sql, $upstream, 0, $child->feature_id);
}
// for internal sub features we don't want upstream or downstream bases
else {
$sql = "EXECUTE sequence_by_parent (%d, %d, %d)";
$q = chado_query($sql, 0, 0, $child->feature_id);
}
while ($subseq = db_fetch_object($q)) {
// concatenate the sequences of all the sub features
if ($subseq->srcfeature_id == $parent->srcfeature_id) {
$seq .= $subseq->residues;
}
}
$i++;
}
}
// if this isn't an aggregate then use the parent residues
else {
$seq = $parent->residues;
}
// get the reverse compliment if feature is on the reverse strand
$dir = 'forward';
if ($parent->strand < 0) {
$seq = tripal_feature_reverse_complement($seq);
$dir = 'reverse';
}
// now format for display
if ($output_format == 'fasta_html') {
$seq = wordwrap($seq, $num_bases_per_line, "
", TRUE);
}
elseif ($output_format == 'fasta_txt') {
$seq = wordwrap($seq, $num_bases_per_line, "\r\n", TRUE);
}
$residues .= ">$feature_name. Sequence derived from feature of type, '$parent->srctypename', of $parent->genus $parent->species: $parent->srcname:" . ($parent->adjfmin + 1) . ".." . $parent->adjfmax ." ($dir). ";
if (count($types) > 0) {
$residues .= "Excludes all bases but those of type(s): " . implode(', ', $types) . ". " ;
}
if ($parent->upstream > 0) {
$residues .= "Includes " . $parent->upstream . " bases upstream. ";
}
if ($parent->downstream > 0) {
$residues .= "Includes " . $parent->downstream . " bases downstream. ";
}
if (!$seq) {
if ($output_format == 'fasta_html') {
$residues .= "No sequence available.";
}
else {
$residues .= "No sequence available.\r\n";
}
}
else {
if ($output_format == 'fasta_html') {
$residues .= "
";
}
$residues .= "\r\n" . $seq . "\r\n";
if ($output_format == 'fasta_html') {
$residues .= "
";
}
}
}
}
// if we are not getting the sequence from the parent sequence then
// use what comes through from the feature record
else {
$sql = "SELECT * FROM {feature} F WHERE feature_id = %d";
$values = db_fetch_object(chado_query($sql, $feature_id));
$residues = $values->residues;
if ($output_format == 'fasta_html') {
$residues = wordwrap($residues, $num_bases_per_line, "
", TRUE);
}
elseif ($output_format == 'fasta_txt') {
$residues = wordwrap($residues, $num_bases_per_line, "\r\n", TRUE);
}
$residues = ">$feature_name\r\n$residues\r\n";
}
// format the residues for display
if ($residues and $num_bases_per_line) {
if ($output_format == 'fasta_html') {
$residues = '' . $residues . '';
}
}
return $residues;
}
/**
* This function defines the custom tables that will be created
* in the chado schema.
*
* @ingroup tripal_feature
*/
function tripal_feature_get_custom_tables($table = NULL) {
if (!$table or strcmp($table, 'tripal_gff_temp')==0) {
$schema['tripal_gff_temp'] = array(
'table' => 'tripal_gff_temp',
'fields' => array(
'feature_id' => array(
'type' => 'int',
'not null' => TRUE,
),
'organism_id' => array(
'type' => 'int',
'not null' => TRUE,
),
'uniquename' => array(
'type' => 'text',
'not null' => TRUE,
),
'type_name' => array(
'type' => 'varchar',
'length' => '1024',
'not null' => TRUE,
),
),
'indexes' => array(
'tripal_gff_temp_idx0' => array('feature_id'),
'tripal_gff_temp_idx0' => array('organism_id'),
'tripal_gff_temp_idx1' => array('uniquename'),
),
'unique keys' => array(
'tripal_gff_temp_uq0' => array('feature_id'),
'tripal_gff_temp_uq1' => array('uniquename', 'organism_id', 'type_name'),
),
);
}
return $schema;
}
/**
* Using the tripal_core_expand_chado_vars function to retrieve a set
* of relationships can be very slow, especialy if there are many relationships
* This function is intended to help speed up the retrieval of relationships
* by only retrieving the base information for the relationship and returning
* an array with
*
* @param $feature
* The feature object
* @return
* An array with two objects
*
* @ingroup tripal_feature_api
*/
function tripal_feature_get_feature_relationships($feature) {
// expand the feature object to include the feature relationships.
$options = array(
'return_array' => 1,
'order_by' => array('rank' => 'ASC'),
);
$feature = tripal_core_expand_chado_vars($feature, 'table',
'feature_relationship', $options);
// get the subject relationships
$srelationships = $feature->feature_relationship->subject_id;
$orelationships = $feature->feature_relationship->object_id;
// get alignment as child. The $feature->featureloc element
// is already populated from the alignment preprocess function
$feature = tripal_core_expand_chado_vars($feature, 'table', 'featureloc');
$cfeaturelocs = $feature->featureloc->feature_id;
if (!$cfeaturelocs) {
$cfeaturelocs = array();
}
elseif (!is_array($cfeaturelocs)) {
$cfeaturelocs = array($cfeaturelocs);
}
// prepare the SQL statement to get the featureloc for the
// feature in the relationships.
$connection = tripal_db_persistent_chado();
$psql = "
PREPARE sel_featureloc_preprocess_relationships (int, int) AS
SELECT
FL.featureloc_id, F.name as srcfeature_name, FL.srcfeature_id,
FL.feature_id, FL.fmin, FL.fmax, FL.strand, FL.phase
FROM featureloc FL
INNER JOIN feature F ON F.feature_id = FL.srcfeature_id
WHERE FL.feature_id = $1 and FL.srcfeature_id = $2
";
tripal_core_chado_prepare('sel_featureloc_preprocess_relationships', $psql, array('int', 'int'));
// combine both object and subject relationshisp into a single array
$relationships = array();
$relationships['object'] = array();
$relationships['subject'] = array();
// iterate through the object relationships
if ($orelationships) {
foreach ($orelationships as $relationship) {
$rel = new stdClass();
// get locations where the child feature and this feature overlap with the
// same landmark feature.
$rel->child_featurelocs = array();
foreach ($cfeaturelocs as $featureloc) {
$res = chado_query("EXECUTE sel_featureloc_preprocess_relationships (%d, %d)",
$relationship->subject_id->feature_id,
$featureloc->srcfeature_id->feature_id);
while ($loc = db_fetch_object($res)) {
// add in the node id of the src feature if it exists and save this location
$loc->nid = $featureloc->srcfeature_id->nid;
$rel->child_featurelocs[] = $loc;
}
}
$rel->record = $relationship;
// get the relationship and child types
$rel_type = t(preg_replace('/_/', " ", $relationship->type_id->name));
$child_type = $relationship->subject_id->type_id->name;
// get the node id of the subject
$sql = "SELECT nid FROM {chado_feature} WHERE feature_id = %d";
$n = db_fetch_object(db_query($sql, $relationship->subject_id->feature_id));
if ($n) {
$rel->record->nid = $n->nid;
}
if (!array_key_exists($rel_type, $relationships['object'])) {
$relationships['object'][$rel_type] = array();
}
if (!array_key_exists($child_type, $relationships['object'][$rel_type])) {
$relationships['object'][$rel_type][$child_type] = array();
}
$relationships['object'][$rel_type][$child_type][] = $rel;
}
}
// now add in the subject relationships
if ($srelationships) {
foreach ($srelationships as $relationship) {
$rel = new stdClass();
// get locations where this feature overlaps with the parent
$rel->parent_featurelocs = array();
foreach ($cfeaturelocs as $featureloc) {
$res = chado_query("EXECUTE sel_featureloc_preprocess_relationships (%d, %d)",
$relationship->object_id->feature_id,
$featureloc->srcfeature_id->feature_id);
while ($loc = db_fetch_object($res)) {
// add in the node id of the src feature if it exists and save this location
$loc->nid = $featureloc->srcfeature_id->nid;
$rel->parent_featurelocs[] = $loc;
}
}
$rel->record = $relationship;
$rel_type = t(preg_replace('/_/', " ", $relationship->type_id->name));
$parent_type = $relationship->object_id->type_id->name;
// get the node id of the subject
$sql = "SELECT nid FROM {chado_feature} WHERE feature_id = %d";
$n = db_fetch_object(db_query($sql, $relationship->object_id->feature_id));
if ($n) {
$rel->record->nid = $n->nid;
}
if (!array_key_exists($rel_type, $relationships['subject'])) {
$relationships['subject'][$rel_type] = array();
}
if (!array_key_exists($child_type, $relationships['subject'][$rel_type])) {
$relationships['subject'][$rel_type][$parent_type] = array();
}
$relationships['subject'][$rel_type][$parent_type][] = $rel;
}
}
return $relationships;
}
/*
* This function adds an entry to the feature_dbxref table.
*
* @param $feature_id
* The numeric feature_if of the feature
* @param $dbname
* The name of the database to which the term belongs
* @param accession
* The accession of the term
*
* @return
* TRUE on success. FALSE on failure.
*
* @ingroup tripal_feature_api
*/
function tripal_feature_add_dbxref($feature_id, $dbname, $accession) {
// make sure the db exists. If it doesn't, then add it
$values = array('name' => $dbname);
$options = array('statement_name' => 'sel_db_na');
$db = tripal_core_chado_select('db', array('db_id'), $values, $options);
if (!$db or count($db) == 0) {
$options = array('statement_name' => 'ins_db_na');
$success = tripal_core_chado_insert('db', $values, $options);
if (!$success) {
watchdog('tripal_feature', 'tripal_feature_add_dbxref: The feature dbxref entry for feature, %feature_id, ".
"could not be added because the database, %dbname, does not exist and cannot be added.',
array('%feature_id' => $feature_id, '%dbname' => $dbname), WATCHDOG_WARNING);
return FALSE;
}
}
// first make sure that the record doesn't already exist
$values = array(
'dbxref_id' => array(
'accession' => $accession,
'db_id' => array(
'name' => $dbname
),
),
'feature_id' => $feature_id,
);
$options = array('statement_name' => 'sel_featuredbxref_dbfe');
$xref = tripal_core_chado_select('feature_dbxref', array('feature_dbxref_id'), $values, $options);
if (count($xref) == 0) {
// if the record doesn't exist then add it.
$options = array('statement_name' => 'ins_featuredbxref_dbfe');
$success = tripal_core_chado_insert('feature_dbxref', $values, $options);
if (!$success) {
watchdog('tripal_feature', 'tripal_feature_add_dbxref: The feature dbxref entry for feature, %feature_id, '.
'could not be added: %db:%accession.', array('%feature_id' => $feature_id, '%db' => $dbname,
'%accession' => $accession), WATCHDOG_WARNING);
return FALSE;
}
}
return TRUE;
}
/*
* This function adds an entry to the feature_cvterm table.
*
* @param $feature_id
* The numeric feature_if of the feature
* @param $cvname
* The name of the controlled vocabulary to which the term belongs
* @param cvterm
* The name of the cvterm
*
* @return
* TRUE on success. FALSE on failure.
*
* @ingroup tripal_feature_api
*/
function tripal_feature_add_cvterm($feature_id, $cvname, $cvterm) {
// make sure the cv exists. If it doesn't, then add it
$values = array('name' => $cvname);
$options = array('statement_name' => 'sel_cv_na');
$cv = tripal_core_chado_select('cv', array('cv_id'), $values, $options);
if (!$cv or count($cv) == 0) {
$options = array('statement_name' => 'ins_cv_na');
$success = tripal_core_chado_insert('cv', $values, $options);
if (!$success) {
watchdog('tripal_feature', 'tripal_feature_add_cvterm: The feature cvterm entry for feature, %feature_id, ".
"could not be added because the CV, %cvname, does not exist and cannot be added.',
array('%feature_id' => $feature_id, '%cvname' => $cvname), WATCHDOG_WARNING);
return FALSE;
}
}
// first make sure that the record doesn't already exist
$values = array(
'cvterm_id' => array(
'name' => $cvterm,
'cv_id' => array(
'name' => $cvname
),
),
'feature_id' => $feature_id,
'pub_id' => 1,
);
$options = array('statement_name' => 'sel_featuredcvterm_cvfepu');
$xref = tripal_core_chado_select('feature_cvterm', array('feature_cvterm_id'), $values, $options);
if (count($xref) == 0) {
// if the record doesn't exist then add it.
$options = array('statement_name' => 'ins_featurecvterm_cvfepu');
$success = tripal_core_chado_insert('feature_cvterm', $values, $options);
if (!$success) {
watchdog('tripal_feature', 'tripal_feature_add_cvterm: The feature cvterm entry for feature, %feature_id, '.
'could not be added: %cvterm.', array('%feature_id' => $feature_id, '%cvterm' => $cvterm), WATCHDOG_WARNING);
return FALSE;
}
}
return TRUE;
}