|
@@ -0,0 +1,921 @@
|
|
|
+<?php
|
|
|
+
|
|
|
+class TaxonomyImporter extends TripalImporter {
|
|
|
+ /**
|
|
|
+ * The name of this loader. This name will be presented to the site
|
|
|
+ * user.
|
|
|
+ */
|
|
|
+ public static $name = 'Chado NCBI Taxonomy Loader';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * The machine name for this loader. This name will be used to construct
|
|
|
+ * the URL for the loader.
|
|
|
+ */
|
|
|
+ public static $machine_name = 'chado_taxonomy';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * A brief description for this loader. This description will be
|
|
|
+ * presented to the site user.
|
|
|
+ */
|
|
|
+ public static $description = 'Imports new organisms from NCBI using taxonomy IDs, or loads taxonomic details about existing organisms.';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * An array containing the extensions of allowed file types.
|
|
|
+ */
|
|
|
+ public static $file_types = array();
|
|
|
+
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Provides information to the user about the file upload. Typically this
|
|
|
+ * may include a description of the file types allowed.
|
|
|
+ */
|
|
|
+ public static $upload_description = '';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * The title that should appear above the upload button.
|
|
|
+ */
|
|
|
+ public static $upload_title = 'File Upload';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * If the loader should require an analysis record. To maintain provenance
|
|
|
+ * we should always indiate where the data we are uploading comes from.
|
|
|
+ * The method that Tripal attempts to use for this by associating upload files
|
|
|
+ * with an analysis record. The analysis record provides the details for
|
|
|
+ * how the file was created or obtained. Set this to FALSE if the loader
|
|
|
+ * should not require an analysis when loading. if $use_analysis is set to
|
|
|
+ * true then the form values will have an 'analysis_id' key in the $form_state
|
|
|
+ * array on submitted forms.
|
|
|
+ */
|
|
|
+ public static $use_analysis = FALSE;
|
|
|
+
|
|
|
+ /**
|
|
|
+ * If the $use_analysis value is set above then this value indicates if the
|
|
|
+ * analysis should be required.
|
|
|
+ */
|
|
|
+ public static $require_analysis = FALSE;
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Text that should appear on the button at the bottom of the importer
|
|
|
+ * form.
|
|
|
+ */
|
|
|
+ public static $button_text = 'Import from NCBI Taxonomy';
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Indicates the methods that the file uploader will support.
|
|
|
+ */
|
|
|
+ public static $methods = array(
|
|
|
+ // Allow the user to upload a file to the server.
|
|
|
+ 'file_upload' => FALSE,
|
|
|
+ // Allow the user to provide the path on the Tripal server for the file.
|
|
|
+ 'file_local' => FALSE,
|
|
|
+ // Allow the user to provide a remote URL for the file.
|
|
|
+ 'file_remote' => FALSE,
|
|
|
+ );
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Indicates if the file must be provided. An example when it may not be
|
|
|
+ * necessary to require that the user provide a file for uploading if the
|
|
|
+ * loader keeps track of previous files and makes those available for
|
|
|
+ * selection.
|
|
|
+ */
|
|
|
+ public static $file_required = FALSE;
|
|
|
+
|
|
|
+
|
|
|
+ /**
|
|
|
+ * The array of arguments used for this loader. Each argument should
|
|
|
+ * be a separate array containing a machine_name, name, and description
|
|
|
+ * keys. This information is used to build the help text for the loader.
|
|
|
+ */
|
|
|
+ public static $argument_list = array();
|
|
|
+
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Indicates how many files are allowed to be uploaded. By default this is
|
|
|
+ * set to allow only one file. Change to any positive number. A value of
|
|
|
+ * zero indicates an unlimited number of uploaded files are allowed.
|
|
|
+ */
|
|
|
+ public static $cardinality = 0;
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Holds the list of all orgainsms currently in Chado. This list
|
|
|
+ * is needed when checking to see if an organism has already been
|
|
|
+ * loaded.
|
|
|
+ */
|
|
|
+ private $all_orgs = array();
|
|
|
+
|
|
|
+ /**
|
|
|
+ * The record from the Chado phylotree table that refers to this
|
|
|
+ * Taxonomic tree.
|
|
|
+ */
|
|
|
+ private $phylotree = NULL;
|
|
|
+
|
|
|
+ /**
|
|
|
+ * The temporary tree array used by the Tripal Phylotree API for
|
|
|
+ * importing a new tree.
|
|
|
+ */
|
|
|
+ private $tree = NULL;
|
|
|
+
|
|
|
+ /**
|
|
|
+ * @see TripalImporter::form()
|
|
|
+ */
|
|
|
+ public function form($form, &$form_state) {
|
|
|
+
|
|
|
+ $form['instructions'] = array(
|
|
|
+ '#type' => 'fieldset',
|
|
|
+ '#title' => 'instructions',
|
|
|
+ '#description' => t('This form is used to import species from the NCBI
|
|
|
+ Taxonomy database into this site. Alternatively, it can import details
|
|
|
+ about organisms from the NCBI Taxonomy database for organisms that
|
|
|
+ already exist on this site. This loader will also construct
|
|
|
+ the taxonomic tree for the species loaded.'),
|
|
|
+ );
|
|
|
+ $form['taxonomy_ids'] = array(
|
|
|
+ '#type' => 'textarea',
|
|
|
+ '#title' => 'Taxonomy ID',
|
|
|
+ '#description' => t('Please provide a list of NCBI taxonomy IDs separated
|
|
|
+ by spaces, tabs or new lines.
|
|
|
+ The information about these organisms will be downloaded and new organism
|
|
|
+ records will be added to this site.')
|
|
|
+ );
|
|
|
+
|
|
|
+ $form['import_existing'] = array(
|
|
|
+ '#type' => 'checkbox',
|
|
|
+ '#title' => 'Import details for existing species.',
|
|
|
+ '#description' => t('The NCBI Taxonomic Importer examines the organisms
|
|
|
+ currently present in the database and queries NCBI for the
|
|
|
+ taxonomic details. If the importer is able to match the
|
|
|
+ genus and species with NCBI the species details will be imported,
|
|
|
+ and a page containing the taxonomic tree will be created.'),
|
|
|
+ '#default value' => 1,
|
|
|
+ );
|
|
|
+ return $form;
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * @see TripalImporter::formValidate()
|
|
|
+ */
|
|
|
+ public function formValidate($form, &$form_state) {
|
|
|
+ global $user;
|
|
|
+
|
|
|
+ $import_existing = $form_state['values']['import_existing'];
|
|
|
+ $taxonomy_ids = $form_state['values']['taxonomy_ids'];
|
|
|
+
|
|
|
+ // make sure that we have numeric values, one per line.
|
|
|
+ if ($taxonomy_ids) {
|
|
|
+ $tax_ids = preg_split("/[\s\n\t\r]+/", $taxonomy_ids);
|
|
|
+ $bad_ids = array();
|
|
|
+ foreach ($tax_ids as $tax_id) {
|
|
|
+ $tax_id = trim($tax_id);
|
|
|
+ if (!preg_match('/^\d+$/', $tax_id)) {
|
|
|
+ $bad_ids[] = $tax_id;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ if (count($bad_ids) > 0) {
|
|
|
+ form_set_error('taxonomy_ids',
|
|
|
+ t('Taxonomy IDs must be numeric. The following are not valid: "@ids".',
|
|
|
+ array('@ids' => implode('", "', $bad_ids))));
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Performs the import.
|
|
|
+ */
|
|
|
+ public function run() {
|
|
|
+ global $site_name;
|
|
|
+
|
|
|
+
|
|
|
+ $arguments = $this->arguments['run_args'];
|
|
|
+ $taxonomy_ids = $arguments['taxonomy_ids'];
|
|
|
+ $import_existing = $arguments['import_existing'];
|
|
|
+
|
|
|
+ // Get the list of all organisms as we'll need this to lookup existing
|
|
|
+ // organisms.
|
|
|
+ if (chado_get_version() > 1.2) {
|
|
|
+ $sql = "
|
|
|
+ SELECT O.*, CVT.name as type
|
|
|
+ FROM {organism} O
|
|
|
+ LEFT JOIN {cvterm} CVT ON CVT.cvterm_id = O.type_id
|
|
|
+ ORDER BY O.genus, O.species
|
|
|
+ ";
|
|
|
+ }
|
|
|
+ else {
|
|
|
+ $sql = "
|
|
|
+ SELECT O.*, '' as type
|
|
|
+ FROM {organism} O
|
|
|
+ ORDER BY O.genus, O.species
|
|
|
+ ";
|
|
|
+ }
|
|
|
+ $results = chado_query($sql);
|
|
|
+ while ($item = $results->fetchObject()) {
|
|
|
+ $this->all_orgs[] = $item;
|
|
|
+ }
|
|
|
+
|
|
|
+ // Get the phylotree object.
|
|
|
+ $this->logMessage('Initializing Tree...');
|
|
|
+ $this->phylotree = $this->initTree();
|
|
|
+ $this->logMessage('Rebuilding Tree...');
|
|
|
+ $this->tree = $this->rebuildTree();
|
|
|
+
|
|
|
+ // Clean out the phnylondes for this tree in the event this is a reload
|
|
|
+ chado_delete_record('phylonode', array('phylotree_id' => $this->phylotree->phylotree_id));
|
|
|
+
|
|
|
+ // Get the taxonomy IDs provided by the user (if any).
|
|
|
+ $tax_ids = array();
|
|
|
+ if ($taxonomy_ids) {
|
|
|
+ $tax_ids = preg_split("/[\s\n\t\r]+/", $taxonomy_ids);
|
|
|
+ }
|
|
|
+
|
|
|
+ // Set the number of items to handle.
|
|
|
+ if ($taxonomy_ids and $import_existing) {
|
|
|
+ $this->setTotalItems(count($this->all_orgs) + count($tax_ids));
|
|
|
+ }
|
|
|
+ if ($taxonomy_ids and !$import_existing) {
|
|
|
+ $this->setTotalItems(count($tax_ids));
|
|
|
+ }
|
|
|
+ if (!$taxonomy_ids and $import_existing) {
|
|
|
+ $this->setTotalItems(count($this->all_orgs));
|
|
|
+ }
|
|
|
+ $this->setItemsHandled(0);
|
|
|
+
|
|
|
+ // If the user wants to import new taxonomy IDs then do that.
|
|
|
+ if ($taxonomy_ids){
|
|
|
+ $this->logMessage('Importing Taxonomy IDs...');
|
|
|
+ foreach ($tax_ids as $tax_id) {
|
|
|
+ $tax_id = trim($tax_id);
|
|
|
+ $this->importRecord($tax_id);
|
|
|
+ $this->addItemsHandled(1);
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ // If the user wants to update existing records then do that.
|
|
|
+ if ($import_existing) {
|
|
|
+ $this->logMessage('Updating Existing...');
|
|
|
+ $this->updateExisting();
|
|
|
+ }
|
|
|
+
|
|
|
+ // Now import the tree.
|
|
|
+ $options = array('taxonomy' => 1);
|
|
|
+ chado_phylogeny_import_tree($this->tree, $this->phylotree, $options);
|
|
|
+ }
|
|
|
+
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Create the taxonomic tree in Chado.
|
|
|
+ *
|
|
|
+ * If the tree already exists it will not be recreated.
|
|
|
+ *
|
|
|
+ * @throws Exception
|
|
|
+ * @return
|
|
|
+ * Returns the phylotree object.
|
|
|
+ */
|
|
|
+ private function initTree() {
|
|
|
+ // Add the taxonomy tree record into the phylotree table. If the tree
|
|
|
+ // already exists then don't insert it again.
|
|
|
+ $site_name = variable_get('site_name');
|
|
|
+ $tree_name = $site_name . 'Taxonomy Tree';
|
|
|
+ $phylotree = chado_select_record('phylotree', array('*'), array('name' => $tree_name));
|
|
|
+ if (count($phylotree) == 0) {
|
|
|
+ // Add the taxonomic tree.
|
|
|
+ $phylotree = array(
|
|
|
+ 'name' => $site_name . 'Taxonomy Tree',
|
|
|
+ 'description' => 'A phylogenetic tree based on taxonomic rank.',
|
|
|
+ 'leaf_type' => 'taxonomy',
|
|
|
+ 'tree_file' => '/dev/null',
|
|
|
+ 'format' => 'taxonomy',
|
|
|
+ 'no_load' => TRUE,
|
|
|
+ );
|
|
|
+ $errors = array();
|
|
|
+ $warnings = array();
|
|
|
+ $success = tripal_insert_phylotree($phylotree, $errors, $warnings);
|
|
|
+ if (!$success) {
|
|
|
+ throw new Exception("Cannot add the Taxonomy Tree record.");
|
|
|
+ }
|
|
|
+ $phylotree = (object) $phylotree;
|
|
|
+ }
|
|
|
+ else {
|
|
|
+ $phylotree = $phylotree[0];
|
|
|
+ }
|
|
|
+ return $phylotree;
|
|
|
+ }
|
|
|
+
|
|
|
+
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Iterates through all existing organisms and rebuilds the taxonomy tree.
|
|
|
+ *
|
|
|
+ * The phloytree API doesn't support adding nodes to existing trees only
|
|
|
+ * importing whole trees. So, we must rebuild the tree using the current
|
|
|
+ * organisms and then we can add to it.
|
|
|
+ *
|
|
|
+ */
|
|
|
+ private function rebuildTree() {
|
|
|
+ $lineage_nodes[] = array();
|
|
|
+
|
|
|
+ // Get the "rank" cvterm. It requires that the TAXRANK vocabulary is loaded.
|
|
|
+ $rank_cvterm = chado_get_cvterm(array(
|
|
|
+ 'name' => 'rank',
|
|
|
+ 'cv_id' => array('name' => 'local')
|
|
|
+ ));
|
|
|
+
|
|
|
+ // The taxonomic tree must have a root, so create that first.
|
|
|
+ $tree = array(
|
|
|
+ 'name' => 'root',
|
|
|
+ 'depth' => 0,
|
|
|
+ 'is_root' => 1,
|
|
|
+ 'is_leaf' => 0,
|
|
|
+ 'is_internal' => 0,
|
|
|
+ 'left_index' => 0,
|
|
|
+ 'right_index' => 0,
|
|
|
+ 'branch_set' => array(),
|
|
|
+ );
|
|
|
+
|
|
|
+ $total = count($this->all_orgs);
|
|
|
+ $j = 1;
|
|
|
+ foreach ($this->all_orgs as $organism) {
|
|
|
+ $sci_name = chado_get_organism_scientific_name($organism);
|
|
|
+ //$this->logMessage("- " . ($j++) . " of $total. Adding @organism", array('@organism' => $sci_name));
|
|
|
+
|
|
|
+ // First get the phylonode record for this organism.
|
|
|
+ $sql = "
|
|
|
+ SELECT P.*
|
|
|
+ FROM {phylonode} P
|
|
|
+ INNER JOIN {phylonode_organism} PO on PO.phylonode_id = P.phylonode_id
|
|
|
+ WHERE P.phylotree_id = :phylotree_id AND PO.organism_id = :organism_id
|
|
|
+ ";
|
|
|
+ $args = array(
|
|
|
+ ':phylotree_id' => $this->phylotree->phylotree_id,
|
|
|
+ ':organism_id' => $organism->organism_id,
|
|
|
+ );
|
|
|
+ $result = chado_query($sql, $args);
|
|
|
+ if (!$result) {
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+ $phylonode = $result->fetchObject();
|
|
|
+
|
|
|
+ // Next get the lineage for this organism.
|
|
|
+ $lineage = $this->getProperty($organism->organism_id, 'lineage');
|
|
|
+ if (!$lineage) {
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+ $lineage_depth = preg_split('/;\s*/', $lineage->value);
|
|
|
+
|
|
|
+ // Now rebuild the tree by first creating the nodes for the full
|
|
|
+ // lineage and then adding the organism as a leaf node.
|
|
|
+ $parent = $tree;
|
|
|
+ $i = 1;
|
|
|
+ $lineage_good = TRUE;
|
|
|
+ foreach ($lineage_depth as $child) {
|
|
|
+ // We need to find the node in the phylotree for this level of the
|
|
|
+ // lineage, but there's a lot of repeats and we don't want to keep
|
|
|
+ // doing the same queries over and over, so we store the nodes
|
|
|
+ // we've already seen in the $lineage_nodes array for fast lookup.
|
|
|
+ if (array_key_exists($child, $lineage_nodes)) {
|
|
|
+ $phylonode = $lineage_nodes[$child];
|
|
|
+ if (!$phylonode) {
|
|
|
+ $lineage_good = FALSE;
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ else {
|
|
|
+ $values = array(
|
|
|
+ 'phylotree_id' => $this->phylotree->phylotree_id,
|
|
|
+ 'label' => $child,
|
|
|
+ );
|
|
|
+ $columns = array('*');
|
|
|
+ $phylonode = chado_select_record('phylonode', $columns, $values);
|
|
|
+ if (count($phylonode) == 0) {
|
|
|
+ $lineage_nodes[$child] = NULL;
|
|
|
+ $lineage_good = FALSE;
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+ $phylonode = $phylonode[0];
|
|
|
+ $lineage_nodes[$child] = $phylonode;
|
|
|
+
|
|
|
+
|
|
|
+ $values = array(
|
|
|
+ 'phylonode_id' => $phylonode->phylonode_id,
|
|
|
+ 'type_id' => $rank_cvterm->cvterm_id,
|
|
|
+ );
|
|
|
+ $columns = array('*');
|
|
|
+ $phylonodeprop = chado_select_record('phylonodeprop', $columns, $values);
|
|
|
+ }
|
|
|
+ $name = $child;
|
|
|
+ $node_rank = (string) $child->Rank;
|
|
|
+ $node = array(
|
|
|
+ 'name' => $name,
|
|
|
+ 'depth' => $i,
|
|
|
+ 'is_root' => 0,
|
|
|
+ 'is_leaf' => 0,
|
|
|
+ 'is_internal' => 1,
|
|
|
+ 'left_index' => 0,
|
|
|
+ 'right_index' => 0,
|
|
|
+ 'parent' => $parent,
|
|
|
+ 'branch_set' => array(),
|
|
|
+ 'parent' => $parent['name'],
|
|
|
+ 'properties' => array(
|
|
|
+ $rank_cvterm->cvterm_id => $phylonodeprop[0]->value,
|
|
|
+ ),
|
|
|
+ );
|
|
|
+ $parent = $node;
|
|
|
+ $this->addTaxonomyNode($tree, $node, $lineage_depth);
|
|
|
+ $i++;
|
|
|
+ } // end foreach ($lineage_depth as $child) { ...
|
|
|
+
|
|
|
+ // If $stop is set then we had problems setting the lineage so
|
|
|
+ // skip adding the leaf node below.
|
|
|
+ if (!$lineage_good) {
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+
|
|
|
+ $rank_type = 'species';
|
|
|
+ if (property_exists($organism, 'type_id') and $organism->type_id) {
|
|
|
+ $rank_type = $organism->type;
|
|
|
+ }
|
|
|
+
|
|
|
+ // Now add in the leaf node
|
|
|
+ $sci_name = chado_get_organism_scientific_name($organism);
|
|
|
+ $node = array(
|
|
|
+ 'name' => $sci_name,
|
|
|
+ 'depth' => $i,
|
|
|
+ 'is_root' => 0,
|
|
|
+ 'is_leaf' => 1,
|
|
|
+ 'is_internal' => 0,
|
|
|
+ 'left_index' => 0,
|
|
|
+ 'right_index' => 0,
|
|
|
+ 'parent' => $parent['name'],
|
|
|
+ 'organism_id' => $organism->organism_id,
|
|
|
+ 'properties' => array(
|
|
|
+ $rank_cvterm->cvterm_id => $rank_type,
|
|
|
+ ),
|
|
|
+ );
|
|
|
+ $this->addTaxonomyNode($tree, $node, $lineage_depth);
|
|
|
+
|
|
|
+ // Set the indecies for the tree.
|
|
|
+ chado_assign_phylogeny_tree_indices($tree);
|
|
|
+ }
|
|
|
+
|
|
|
+ return $tree;
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Imports details from NCBI Taxonomy for organisms that alrady exist.
|
|
|
+ */
|
|
|
+ private function updateExisting() {
|
|
|
+
|
|
|
+ foreach ($this->all_orgs as $organism) {
|
|
|
+ // If the organism record is marked as new then let's skip it because
|
|
|
+ // it was newly added and should have the updated information already.
|
|
|
+ if ($organism->is_new) {
|
|
|
+ continue;
|
|
|
+ }
|
|
|
+
|
|
|
+ // TODO: we should check if the organism already has a taxonomy ID.
|
|
|
+ // if so we should use that instead of the scientific name.
|
|
|
+
|
|
|
+ // Build the query string to get the information about this species.
|
|
|
+ $sci_name = chado_get_organism_scientific_name($organism);
|
|
|
+ $sci_name = urlencode($sci_name);
|
|
|
+ $search_url = "http://www.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?".
|
|
|
+ "db=taxonomy" .
|
|
|
+ "&term=$sci_name";
|
|
|
+
|
|
|
+ // Get the search response from NCBI.
|
|
|
+ $rfh = fopen($search_url, "r");
|
|
|
+ $xml_text = '';
|
|
|
+ while (!feof($rfh)) {
|
|
|
+ $xml_text .= fread($rfh, 255);
|
|
|
+ }
|
|
|
+ fclose($rfh);
|
|
|
+
|
|
|
+ // Parse the XML to get the taxonomy ID
|
|
|
+ $xml = new SimpleXMLElement($xml_text);
|
|
|
+ if ($xml) {
|
|
|
+ $taxid = (string) $xml->IdList->Id;
|
|
|
+ if ($taxid) {
|
|
|
+ $this->importRecord($taxid, $organism);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ $this->addItemsHandled(1);
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Checks the Chado database to see if the organism already exists.
|
|
|
+ *
|
|
|
+ * @param $taxid
|
|
|
+ * The taxonomic ID for the organism.
|
|
|
+ * @param $sci_name
|
|
|
+ * The scientific name for the organism as returned by NCBI
|
|
|
+ */
|
|
|
+ private function findOrganism($taxid, $sci_name) {
|
|
|
+ $organism = NULL;
|
|
|
+
|
|
|
+ // First check the taxid to see if it's present and assocaited with an
|
|
|
+ // organism already.
|
|
|
+ $values = array(
|
|
|
+ 'db_id' => array(
|
|
|
+ 'name' => 'NCBITaxon'
|
|
|
+ ),
|
|
|
+ 'accession' => $taxid,
|
|
|
+ );
|
|
|
+ $columns = array('dbxref_id');
|
|
|
+ $dbxref = chado_select_record('dbxref', $columns, $values);
|
|
|
+ if (count($dbxref) > 0) {
|
|
|
+ $columns = array('organism_id');
|
|
|
+ $values = array('dbxref_id' => $dbxref[0]->dbxref_id);
|
|
|
+ $organism_dbxref = chado_select_record('organism_dbxref', $columns, $values);
|
|
|
+ if (count($organism_dbxref) >0) {
|
|
|
+ $organism_id = $organism_dbxref[0]->organism_id;
|
|
|
+ $columns = array('*');
|
|
|
+ $values = array('organism_id' => $organism_id);
|
|
|
+ $organism = chado_select_record('organism', $columns, $values);
|
|
|
+ if (count($organism) > 0) {
|
|
|
+ $organism = $organism[0];
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ // If the caller did not provide an organism then we want to try and
|
|
|
+ // add one. But, it only makes sense to add one if this record
|
|
|
+ // is of rank species.
|
|
|
+ // First check if the full name (including the infrasepcific name)
|
|
|
+ // are all present in the genus and species name. This would have
|
|
|
+ // been the Chado v1.2 (or less) of storing species.
|
|
|
+ if (!$organism) {
|
|
|
+ $sql = "
|
|
|
+ SELECT organism_id
|
|
|
+ FROM {organism}
|
|
|
+ WHERE concat(genus, ' ', species) = :sci_name
|
|
|
+ ";
|
|
|
+ $results = chado_query($sql, array(':sci_name' => $sci_name));
|
|
|
+ $item = $results->fetchObject();
|
|
|
+ if ($item) {
|
|
|
+ $columns = array('*');
|
|
|
+ $values = array('organism_id' => $item->organism_id);
|
|
|
+ $organism = chado_select_record('organism', $columns, $values);
|
|
|
+ if (count($organism) > 0) {
|
|
|
+ $organism = $organism[0];
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+ // Second, check if the full name includes the infraspecific name.
|
|
|
+ if (!$organism) {
|
|
|
+ foreach ($this->all_orgs as $item) {
|
|
|
+ $internal_sci_name = chado_get_organism_scientific_name($item);
|
|
|
+ if ($sci_name == $internal_sci_name) {
|
|
|
+ $organism = $item;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+ return $organism;
|
|
|
+ }
|
|
|
+ /**
|
|
|
+ * Adds a new organism record to Chado.
|
|
|
+ *
|
|
|
+ * @param sci_name
|
|
|
+ * The scientific name as provied by NCBI Taxonomy.
|
|
|
+ * @param $rank
|
|
|
+ * The rank of the organism as provied by NCBI Taxonomy.
|
|
|
+ */
|
|
|
+ private function addOrganism($sci_name, $rank) {
|
|
|
+
|
|
|
+ $organism = NULL;
|
|
|
+ $matches = array();
|
|
|
+ $genus = '';
|
|
|
+ $species = '';
|
|
|
+ $infra = '';
|
|
|
+ $values = array();
|
|
|
+
|
|
|
+ // Check if the scientific name has an infraspecific part or is just
|
|
|
+ // a species name.
|
|
|
+ if (preg_match('/^(.+?)\s+(.+?)\s+(.+)$/', $sci_name, $matches)) {
|
|
|
+ $genus = $matches[1];
|
|
|
+ $species = $matches[2];
|
|
|
+ $infra = $matches[3];
|
|
|
+
|
|
|
+ // Get the CV term for the rank.
|
|
|
+ $type = chado_get_cvterm(array(
|
|
|
+ 'name' => preg_replace('/ /','_', $rank),
|
|
|
+ 'cv_id' => array('name' => 'taxonomic_rank')
|
|
|
+ ));
|
|
|
+
|
|
|
+ // Remove the rank from the infraspecific name.
|
|
|
+ $abbrev = chado_abbreviate_infraspecific_rank($rank);
|
|
|
+ $infra = preg_replace("/$abbrev/", "", $infra);
|
|
|
+ $infra = trim($infra);
|
|
|
+
|
|
|
+ $values = array(
|
|
|
+ 'genus' => $genus,
|
|
|
+ 'species' => $species,
|
|
|
+ 'abbreviation' => $genus[0] . '. ' . $species,
|
|
|
+ 'type_id' => $type->cvterm_id,
|
|
|
+ 'infraspecific_name' => $infra,
|
|
|
+ );
|
|
|
+ $organism = chado_insert_record('organism', $values);
|
|
|
+ $organism = (object) $organism;
|
|
|
+ $organism->type = $rank;
|
|
|
+ }
|
|
|
+ else if (preg_match('/^(.+?)\s+(.+?)$/', $sci_name, $matches)) {
|
|
|
+ $genus = $matches[1];
|
|
|
+ $species = $matches[2];
|
|
|
+ $infra = '';
|
|
|
+ $values = array(
|
|
|
+ 'genus' => $genus,
|
|
|
+ 'species' => $species,
|
|
|
+ 'abbreviation' => $genus[0] . '. ' . $species,
|
|
|
+ );
|
|
|
+ $organism = chado_insert_record('organism', $values);
|
|
|
+ $organism = (object) $organism;
|
|
|
+ }
|
|
|
+ if ($organism) {
|
|
|
+ $organism->is_new = TRUE;
|
|
|
+ $this->all_orgs[] = $organism;
|
|
|
+ }
|
|
|
+
|
|
|
+ return $organism;
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Imports an organism from the NCBI taxonomy DB by its taxonomy ID
|
|
|
+ *
|
|
|
+ * @param $taxid
|
|
|
+ * The NCBI Taxonomy ID.
|
|
|
+ * @param $organism
|
|
|
+ * The organism object to which this taxonomy belongs. If the organism
|
|
|
+ * is NULL then it will be created.
|
|
|
+ */
|
|
|
+ private function importRecord($taxid, $organism = NULL) {
|
|
|
+ $adds_organism = $organism ? FALSE : TRUE;
|
|
|
+
|
|
|
+ // Get the "rank" cvterm. It requires that the TAXRANK vocabulary is loaded.
|
|
|
+ $rank_cvterm = chado_get_cvterm(array(
|
|
|
+ 'name' => 'rank',
|
|
|
+ 'cv_id' => array('name' => 'local')
|
|
|
+ ));
|
|
|
+
|
|
|
+ // Get the details for this taxonomy.
|
|
|
+ $fetch_url = "http://www.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?".
|
|
|
+ "db=taxonomy" .
|
|
|
+ "&id=$taxid";
|
|
|
+
|
|
|
+ // Get the search response from NCBI.
|
|
|
+ $rfh = fopen($fetch_url, "r");
|
|
|
+ $xml_text = '';
|
|
|
+ while (!feof($rfh)) {
|
|
|
+ $xml_text .= fread($rfh, 255);
|
|
|
+ }
|
|
|
+ fclose($rfh);
|
|
|
+
|
|
|
+ $xml = new SimpleXMLElement($xml_text);
|
|
|
+ if ($xml) {
|
|
|
+ $taxon = $xml->Taxon;
|
|
|
+
|
|
|
+ // Get the genus and species from the xml.
|
|
|
+ $parent = (string) $taxon->ParentTaxId;
|
|
|
+ $rank = (string) $taxon->Rank;
|
|
|
+ $sci_name = (string) $taxon->ScientificName;
|
|
|
+ //$this->logMessage(' - Importing @sci_name', array('@sci_name' => $sci_name));
|
|
|
+
|
|
|
+ // If we don't have an organism record provided then see if there
|
|
|
+ // is one provided by Chado, if not, the try to add one.
|
|
|
+ if (!$organism) {
|
|
|
+ $organism = $this->findOrganism($taxid, $sci_name);
|
|
|
+ if (!$organism) {
|
|
|
+ $organism = $this->addOrganism($sci_name, $rank);
|
|
|
+ if (!$organism) {
|
|
|
+ throw new Exception(t('Cannot add organism: @sci_name', array('@sci_name' => $sci_name)));
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ // Associate the Dbxref with the organism.
|
|
|
+ $this->addDbxref($organism->organism_id, $taxid);
|
|
|
+
|
|
|
+ // Get properties for this organism.
|
|
|
+ $lineage = (string) $taxon->Lineage;
|
|
|
+ $genetic_code = (string) $taxon->GeneticCode->GCId;
|
|
|
+ $genetic_code_name = (string) $taxon->GeneticCode->GCName;
|
|
|
+ $mito_genetic_code = (string) $taxon->MitoGeneticCode->MGCId;
|
|
|
+ $mito_genetic_code_name = (string) $taxon->MitoGeneticCode->MGCName;
|
|
|
+ $division = (string) $taxon->Division;
|
|
|
+
|
|
|
+ // Add in the organism properties.
|
|
|
+ $this->addProperty($organism->organism_id, 'division', $division);
|
|
|
+ $this->addProperty($organism->organism_id, 'mitochondrial_genetic_code_name', $mito_genetic_code_name);
|
|
|
+ $this->addProperty($organism->organism_id, 'mitochondrial_genetic_code', $mito_genetic_code);
|
|
|
+ $this->addProperty($organism->organism_id, 'genetic_code_name', $genetic_code_name);
|
|
|
+ $this->addProperty($organism->organism_id, 'lineage', $lineage);
|
|
|
+ $this->addProperty($organism->organism_id, 'genetic_code', $genetic_code);
|
|
|
+
|
|
|
+ $name_ranks = array();
|
|
|
+ if ($taxon->OtherNames->children) {
|
|
|
+ foreach ($taxon->OtherNames->children() as $child) {
|
|
|
+ $type = $child->getName();
|
|
|
+ $name = (string) $child;
|
|
|
+ if (!array_key_exists($type, $name_ranks)) {
|
|
|
+ $name_ranks[$type] = 0;
|
|
|
+ }
|
|
|
+ switch ($type) {
|
|
|
+ case 'GenbankCommonName':
|
|
|
+ $this->addProperty($organism->organism_id, 'genbank_common_name', $name, $name_ranks[$type]);
|
|
|
+ break;
|
|
|
+ case 'Synonym':
|
|
|
+ case 'GenbankSynonym':
|
|
|
+ $this->addProperty($organism->organism_id, 'synonym', $name, $name_ranks[$type]);
|
|
|
+ break;
|
|
|
+ case 'CommonName':
|
|
|
+ // If we had to add the organism then include the commone name too.
|
|
|
+ if ($adds_organism) {
|
|
|
+ $organism->common_name = $name;
|
|
|
+ $values = array('organism_id' => $organism->id);
|
|
|
+ chado_update_record('organism', $values, $organism);
|
|
|
+ }
|
|
|
+ case 'Includes':
|
|
|
+ $this->addProperty($organism->organism_id, 'other_name', $name, $name_ranks[$type]);
|
|
|
+ break;
|
|
|
+ case 'EquivalentName':
|
|
|
+ $this->addProperty($organism->organism_id, 'equivalent_name', $name, $name_ranks[$type]);
|
|
|
+ break;
|
|
|
+ case 'Anamorph':
|
|
|
+ $this->addProperty($organism->organism_id, 'anamorph', $name, $name_ranks[$type]);
|
|
|
+ break;
|
|
|
+ case 'Name':
|
|
|
+ // skip the Name stanza
|
|
|
+ break;
|
|
|
+ default:
|
|
|
+ print "NOTICE: Skipping unrecognzed name type: $type\n";
|
|
|
+ // do nothing for unrecognized types
|
|
|
+ }
|
|
|
+ $name_ranks[$type]++;
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ // Generate a nested array structure that can be used for importing the tree.
|
|
|
+ $lineage_depth = preg_split('/;\s*/', $lineage);
|
|
|
+ $parent = $this->tree;
|
|
|
+ $i = 1;
|
|
|
+ foreach ($taxon->LineageEx->children() as $child) {
|
|
|
+ $tid = (string) $child->TaxID;
|
|
|
+ $name = (string) $child->ScientificName;
|
|
|
+ $node_rank = (string) $child->Rank;
|
|
|
+ $node = array(
|
|
|
+ 'name' => $name,
|
|
|
+ 'depth' => $i,
|
|
|
+ 'is_root' => 0,
|
|
|
+ 'is_leaf' => 0,
|
|
|
+ 'is_internal' => 1,
|
|
|
+ 'left_index' => 0,
|
|
|
+ 'right_index' => 0,
|
|
|
+ 'parent' => $parent,
|
|
|
+ 'branch_set' => array(),
|
|
|
+ 'parent' => $parent['name'],
|
|
|
+ 'properties' => array(
|
|
|
+ $rank_cvterm->cvterm_id => $node_rank,
|
|
|
+ ),
|
|
|
+ );
|
|
|
+ $parent = $node;
|
|
|
+ $this->addTaxonomyNode($this->tree, $node, $lineage_depth);
|
|
|
+ $i++;
|
|
|
+ }
|
|
|
+ // Now add in the leaf node
|
|
|
+ $node = array(
|
|
|
+ 'name' => $sci_name,
|
|
|
+ 'depth' => $i,
|
|
|
+ 'is_root' => 0,
|
|
|
+ 'is_leaf' => 1,
|
|
|
+ 'is_internal' => 0,
|
|
|
+ 'left_index' => 0,
|
|
|
+ 'right_index' => 0,
|
|
|
+ 'parent' => $parent['name'],
|
|
|
+ 'organism_id' => $organism->organism_id,
|
|
|
+ 'properties' => array(
|
|
|
+ $rank_cvterm->cvterm_id => $rank,
|
|
|
+ ),
|
|
|
+ );
|
|
|
+ $this->addTaxonomyNode($this->tree, $node, $lineage_depth);
|
|
|
+
|
|
|
+ // Set the indecies for the tree.
|
|
|
+ chado_assign_phylogeny_tree_indices($this->tree);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ /**
|
|
|
+ *
|
|
|
+ */
|
|
|
+ private function addTaxonomyNode(&$tree, $node, $lineage_depth) {
|
|
|
+
|
|
|
+ // Get the branch set for the tree root.
|
|
|
+ $branch_set = &$tree['branch_set'];
|
|
|
+
|
|
|
+ // Iterate through the tree up until the depth where this node will
|
|
|
+ // be placed.
|
|
|
+ $node_depth = $node['depth'];
|
|
|
+ for ($i = 1; $i <= $node_depth; $i++) {
|
|
|
+ // Iterate through any existing nodes in the branch set to see if
|
|
|
+ // the node name matches the correct name for the lineage at this
|
|
|
+ // depth. If it matches then it is inside of this branch set that
|
|
|
+ // we will place the node.
|
|
|
+ for ($j = 0; $j < count($branch_set); $j++) {
|
|
|
+ // If this node already exists in the tree then return.
|
|
|
+ if ($branch_set[$j]['name'] == $node['name'] and
|
|
|
+ $branch_set[$j]['depth'] = $node['depth']) {
|
|
|
+ return;
|
|
|
+ }
|
|
|
+ // Otherwise, set the branch to be the current branch and continue.
|
|
|
+ if ($branch_set[$j]['name'] == $lineage_depth[$i-1]) {
|
|
|
+ $branch_set = &$branch_set[$j]['branch_set'];
|
|
|
+ break;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+ // Add the node to the last branch set. This should be where this node goes.
|
|
|
+ $branch_set[] = $node;
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Retrieves a property for a given organism.
|
|
|
+ *
|
|
|
+ * @param $organism_id
|
|
|
+ * The organism ID to which the property is added.
|
|
|
+ * @param $term_name
|
|
|
+ * The name of the organism property term. This term must be
|
|
|
+ * present in the 'organism_property' cv.
|
|
|
+ * @param $rank
|
|
|
+ * The order for this property. The first instance of this term for
|
|
|
+ * this organism should be zero. Defaults to zero.
|
|
|
+ * @return
|
|
|
+ * The property object.
|
|
|
+ */
|
|
|
+ private function getProperty($organism_id, $term_name, $rank = 0) {
|
|
|
+ $record = array(
|
|
|
+ 'table' => 'organism',
|
|
|
+ 'id' => $organism_id
|
|
|
+ );
|
|
|
+ $property = array(
|
|
|
+ 'type_name' => $term_name,
|
|
|
+ 'cv_name' => 'organism_property',
|
|
|
+ 'value' => $value,
|
|
|
+ 'rank' => $rank
|
|
|
+ );
|
|
|
+
|
|
|
+ return chado_get_property($record, $property);
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ * Adds a property to an organism node.
|
|
|
+ *
|
|
|
+ * @param $organism_id
|
|
|
+ * The organism ID to which the property is added.
|
|
|
+ * @param $term_name
|
|
|
+ * The name of the organism property term. This term must be
|
|
|
+ * present in the 'organism_property' cv.
|
|
|
+ * @param $value
|
|
|
+ * The value of the property.
|
|
|
+ * @param $rank
|
|
|
+ * The order for this property. The first instance of this term for
|
|
|
+ * this organism should be zero. Defaults to zero.
|
|
|
+ */
|
|
|
+ private function addProperty($organism_id, $term_name, $value, $rank = 0) {
|
|
|
+ if (!$value) {
|
|
|
+ return;
|
|
|
+ }
|
|
|
+
|
|
|
+ $record = array(
|
|
|
+ 'table' => 'organism',
|
|
|
+ 'id' => $organism_id
|
|
|
+ );
|
|
|
+ $property = array(
|
|
|
+ 'type_name' => $term_name,
|
|
|
+ 'cv_name' => 'organism_property',
|
|
|
+ 'value' => $value
|
|
|
+ );
|
|
|
+ // Delete all properties of this type if the rank is zero.
|
|
|
+ if ($rank == 0) {
|
|
|
+ chado_delete_property($record, $property);
|
|
|
+ }
|
|
|
+ chado_insert_property($record, $property);
|
|
|
+ }
|
|
|
+
|
|
|
+ /**
|
|
|
+ *
|
|
|
+ * @param unknown $organism_id
|
|
|
+ * @param unknown $taxId
|
|
|
+ */
|
|
|
+ private function addDbxref($organism_id, $taxId) {
|
|
|
+ $db = chado_get_db(array('name' => 'NCBITaxon'));
|
|
|
+ $values = array(
|
|
|
+ 'db_id' => $db->db_id,
|
|
|
+ 'accession' => $taxId
|
|
|
+ );
|
|
|
+ $dbxref = chado_insert_dbxref($values);
|
|
|
+
|
|
|
+ $values = array(
|
|
|
+ 'dbxref_id' => $dbxref->dbxref_id,
|
|
|
+ 'organism_id' => $organism_id,
|
|
|
+ );
|
|
|
+
|
|
|
+ if (!chado_select_record('organism_dbxref', ['organism_dbxref_id'], $values)) {
|
|
|
+ chado_insert_record('organism_dbxref', $values);
|
|
|
+ }
|
|
|
+ }
|
|
|
+}
|