|
|
@@ -113,7 +113,7 @@ function tripal_analysis_kegg_brite($analysis_id, $type_id, $ajax){
|
|
|
<td nowrap valign=\"top\">
|
|
|
";
|
|
|
// List all BRITE terms on the left
|
|
|
- $sql = "SELECT CVT.name, CVT.cvterm_id
|
|
|
+ $sql = "SELECT DISTINCT CVT.name, CVT.cvterm_id
|
|
|
FROM {cvterm} CVT
|
|
|
INNER JOIN analysisprop AP ON CVT.cvterm_id = AP.type_id
|
|
|
WHERE AP.analysis_id = %d AND CVT.definition LIKE 'KEGG BRITE term: %'
|
|
|
@@ -378,6 +378,14 @@ function tripal_analysis_kegg_parseHierFile ($analysis_id, $hierfile, $base_path
|
|
|
print "There are $total_files keg file(s).\n";
|
|
|
$interval = intval($total_files * 0.01);
|
|
|
$no_file = 0;
|
|
|
+
|
|
|
+ // Remove the analysis features for this analysis
|
|
|
+ // we will rebuild them from just this parsing
|
|
|
+ if(!tripal_core_chado_delete('analysisfeature',array('analysis_id' => $analysis_id))){
|
|
|
+ print "ERROR: Cannot prepare the analysis for addint features\n";
|
|
|
+ exit;
|
|
|
+ }
|
|
|
+
|
|
|
while ($file = readdir($dir_handle)) {
|
|
|
|
|
|
if(preg_match("/^.*\.keg/",$file)){
|
|
|
@@ -390,7 +398,8 @@ function tripal_analysis_kegg_parseHierFile ($analysis_id, $hierfile, $base_path
|
|
|
$no_file ++;
|
|
|
|
|
|
# $type variable will be set in tripal_analysis_kegg_parse_kegg_file()
|
|
|
- $content = tripal_analysis_kegg_parse_kegg_file("$hierdir/$file",$type,$analysis_id, $base_path, $query_re,$query_type,$query_uniquename);
|
|
|
+ $content = tripal_analysis_kegg_parse_kegg_file("$hierdir/$file",$type,
|
|
|
+ $analysis_id, $base_path, $query_re,$query_type,$query_uniquename);
|
|
|
|
|
|
# add the item to the database
|
|
|
if($content){
|
|
|
@@ -460,12 +469,14 @@ function tripal_analysis_kegg_parse_kegg_file ($file,&$type,$analysis_id, $base_
|
|
|
$id = 0;
|
|
|
$type_id = 0;
|
|
|
$no_line = 0;
|
|
|
-
|
|
|
+ $ul_content = array();
|
|
|
+ $has_feature = 0;
|
|
|
+ $level = -1;
|
|
|
|
|
|
while($line = fgets($handle)){
|
|
|
$no_line ++;
|
|
|
// Find out what kind of file we're looking at.
|
|
|
- if(preg_match("/#DEFINITION\s+(.*)\s+.+?$/",$line,$matches)){
|
|
|
+ if(preg_match("/#.*nbsp;\s(.*)<\/h2>$/",$line,$matches)){
|
|
|
// Set type as the matched term in the DEFINITION line and add it as a new cvterm
|
|
|
$type = $matches[1];
|
|
|
// Before inserting, make sure this cvterm doesn't exist
|
|
|
@@ -492,16 +503,33 @@ function tripal_analysis_kegg_parse_kegg_file ($file,&$type,$analysis_id, $base_
|
|
|
}
|
|
|
// get the depth of the hierarch (e.g. A,B,C or D);
|
|
|
preg_match("/^([ABCDEFGHIJKLMNOP])\s*(.*)/",$line,$matches);
|
|
|
- # skip lines that aren't data or are empty
|
|
|
+
|
|
|
+ // skip lines that aren't data or are empty
|
|
|
if(!$matches[1] or !$matches[2]){continue;}
|
|
|
+
|
|
|
+ // set the current level and keep track of the previous level (e.g. A,B,C...)
|
|
|
$prev = $current;
|
|
|
$current = $matches[1];
|
|
|
|
|
|
- for($i = (ord($current) - ord($prev)); $i < 0; $i++){
|
|
|
- $content .= "</li></ul>\n";
|
|
|
+ // if we have matches and we are going down to a lower level then we
|
|
|
+ // want to close off each <ul> until we hit the level we are currently at
|
|
|
+ // and at the same time we want to keep the text. If we don't have a
|
|
|
+ // match then we'll throw out the text.
|
|
|
+ if($has_feature){
|
|
|
+ for($i = (ord($current) - ord($prev)); $i < 0; $i++){
|
|
|
+ $ul_content[$level] .= "</ul>\n";
|
|
|
+ $content .= $ul_content[$level];
|
|
|
+ $ul_content[$level--] = '';
|
|
|
+ }
|
|
|
+ $has_feature = 0;
|
|
|
+ } else {
|
|
|
+ for($i = (ord($current) - ord($prev)); $i < 0; $i++){
|
|
|
+ $ul_content[$level--] = '';
|
|
|
+ }
|
|
|
}
|
|
|
+ // if we've gone up a level then add a new <ul> block
|
|
|
for($i = 0; $i < (ord($current) - ord($prev)); $i++){
|
|
|
- $content .= "<ul>\n";
|
|
|
+ $ul_content[++$level] .= "<ul>\n";
|
|
|
}
|
|
|
|
|
|
// change all relative paths to absolute paths pointing to KEGG (www.genome.jp)
|
|
|
@@ -511,7 +539,9 @@ function tripal_analysis_kegg_parse_kegg_file ($file,&$type,$analysis_id, $base_
|
|
|
$matches[2] = preg_replace("/<a href=\"/i","<a id=\"tripal_kegg_brite_links\" target=\"_blank\" href=\"",$matches[2]);
|
|
|
|
|
|
// extract the features that have been mapped to the KEGG IDs
|
|
|
- if(preg_match("/^(.*?);\s*(\<a.+)/",$matches[2],$defline)){
|
|
|
+ if(preg_match("/^(.*?);\s*(\<a.+)/",$matches[2],$keggline)){
|
|
|
+ $has_feature = 1;
|
|
|
+
|
|
|
// Find cvterm_id for 'kegg_brite_data'
|
|
|
$sql = "SELECT cvterm_id
|
|
|
FROM {cvterm} CVT
|
|
|
@@ -523,17 +553,17 @@ function tripal_analysis_kegg_parse_kegg_file ($file,&$type,$analysis_id, $base_
|
|
|
tripal_db_set_active($previous_db);
|
|
|
|
|
|
// get the feature name using the user's regular expression
|
|
|
- if ($query_re and preg_match("/$query_re/", $defline[1], $parts)) {
|
|
|
+ if ($query_re and preg_match("/$query_re/", $keggline[1], $parts)) {
|
|
|
$feature = $parts[1];
|
|
|
}
|
|
|
// If not in above format then pull up to the first space
|
|
|
else {
|
|
|
- if (preg_match('/^(.*?)\s.*$/', $defline[1], $parts)) {
|
|
|
+ if (preg_match('/^(.*?)\s.*$/', $keggline[1], $parts)) {
|
|
|
$feature = $parts[1];
|
|
|
}
|
|
|
// if no match up to the first space then just use the entire string
|
|
|
else {
|
|
|
- $feature = $defline[1];
|
|
|
+ $feature = $keggline[1];
|
|
|
}
|
|
|
}
|
|
|
|
|
|
@@ -562,66 +592,58 @@ function tripal_analysis_kegg_parse_kegg_file ($file,&$type,$analysis_id, $base_
|
|
|
print "Failed: '$feature' cannot find a matching feature in the database.\n";
|
|
|
continue;
|
|
|
}
|
|
|
- print "Adding KEGG results for $feature ($feature_id,$analysis_id)\n";
|
|
|
+ print "Adding KEGG results for $feature ($feature_id,$analysis_id): $type\n";
|
|
|
$feature_id = $feature_arr[0]->feature_id;
|
|
|
|
|
|
- // get the node ID of the feature if one exists
|
|
|
- $sql = "SELECT nid FROM {chado_feture} WHERE feature_id = %d";
|
|
|
- $nid = db_result(db_query($sql, $feature_id));
|
|
|
-
|
|
|
- // Get the higest rank for this feature_id in analysisfeatureprop table
|
|
|
- $sql = "SELECT MAX(rank) FROM {analysisfeatureprop} AFP ".
|
|
|
- "INNER JOIN analysisfeature AF ON AF.analysisfeature_id = AFP.analysisfeature_id ".
|
|
|
- "WHERE feature_id=%d ".
|
|
|
- "AND analysis_id=%d ".
|
|
|
- "AND type_id=%d ";
|
|
|
- $previous_db = tripal_db_set_active('chado');
|
|
|
- $afp = db_fetch_object(db_query($sql, $feature_id, $analysis_id, $brite_data_type_id));
|
|
|
- tripal_db_set_active($prevous_db);
|
|
|
- $hi_rank = 0;
|
|
|
- if ($afp) {
|
|
|
- $hi_rank = $afp->max + 1;
|
|
|
- }
|
|
|
- //------------------------------------------------------
|
|
|
- // Insert into analysisfeature table
|
|
|
- //------------------------------------------------------
|
|
|
- $sql = "INSERT INTO {analysisfeature} (feature_id, analysis_id) ".
|
|
|
- "VALUES (%d, %d)";
|
|
|
- $previous_db = tripal_db_set_active('chado');
|
|
|
- db_query ($sql, $feature_id, $analysis_id);
|
|
|
- tripal_db_set_active($previous_db);
|
|
|
- // Get the newly inserted analysisfeature_id
|
|
|
- $sql = "SELECT analysisfeature_id FROM {analysisfeature} WHERE feature_id = %d AND analysis_id = %d";
|
|
|
- $previous_db = tripal_db_set_active('chado');
|
|
|
- $analysisfeature_id = db_result(db_query($sql, $feature_id, $analysis_id));
|
|
|
- tripal_db_set_active($previous_db);
|
|
|
- //------------------------------------------------------
|
|
|
- // Insert into analysisfeatureprop table
|
|
|
- //------------------------------------------------------
|
|
|
- // Before inserting, make sure it's not a duplicatefs
|
|
|
- $sql = "SELECT value FROM {analysisfeatureprop} WHERE analysisfeature_id = %d AND type_id = %d";
|
|
|
- $previous_db = tripal_db_set_active('chado');
|
|
|
- $result = db_query($sql, $analysisfeature_id, $brite_data_type_id);
|
|
|
- tripal_db_set_active($previous_db);
|
|
|
- $duplicate = 0;
|
|
|
- while ($afp_value = db_fetch_object($result)) {
|
|
|
- preg_match("/<a.+?>(.+)<\/a>/",$afp_value->value,$old_ids);
|
|
|
- preg_match("/<a.+?>(.+)<\/a>/",$mat[2], $new_ids);
|
|
|
- if ($old_ids[1] && $old_ids[1] == $new_ids[1]) {
|
|
|
- $duplicate = 1;
|
|
|
+ if($feature_id){
|
|
|
+ // get the node ID of the feature if one exists
|
|
|
+ $sql = "SELECT nid FROM {chado_feature} WHERE feature_id = %d";
|
|
|
+ $nid = db_result(db_query($sql, $feature_id));
|
|
|
+
|
|
|
+ //------------------------------------------------------
|
|
|
+ // Insert into analysisfeature table
|
|
|
+ //------------------------------------------------------
|
|
|
+ // Get the higest rank for this feature_id in analysisfeatureprop table
|
|
|
+ $sql = "SELECT MAX(rank) FROM {analysisfeatureprop} AFP ".
|
|
|
+ "INNER JOIN analysisfeature AF ON AF.analysisfeature_id = AFP.analysisfeature_id ".
|
|
|
+ "WHERE feature_id=%d ".
|
|
|
+ "AND analysis_id=%d ".
|
|
|
+ "AND type_id=%d ";
|
|
|
+ $previous_db = tripal_db_set_active('chado');
|
|
|
+ $afp = db_fetch_object(db_query($sql, $feature_id, $analysis_id, $brite_data_type_id));
|
|
|
+ tripal_db_set_active($prevous_db);
|
|
|
+ $hi_rank = 0;
|
|
|
+ if ($afp) {
|
|
|
+ $hi_rank = $afp->max + 1;
|
|
|
}
|
|
|
- }
|
|
|
- if (!$duplicate) {
|
|
|
+
|
|
|
+ $sql = "INSERT INTO {analysisfeature} (feature_id, analysis_id) ".
|
|
|
+ "VALUES (%d, %d)";
|
|
|
+ $previous_db = tripal_db_set_active('chado');
|
|
|
+ db_query ($sql, $feature_id, $analysis_id);
|
|
|
+ tripal_db_set_active($previous_db);
|
|
|
+
|
|
|
+ // Get the newly inserted analysisfeature_id
|
|
|
+ $sql = "SELECT analysisfeature_id FROM {analysisfeature} WHERE feature_id = %d AND analysis_id = %d";
|
|
|
+ $previous_db = tripal_db_set_active('chado');
|
|
|
+ $analysisfeature_id = db_result(db_query($sql, $feature_id, $analysis_id));
|
|
|
+ tripal_db_set_active($previous_db);
|
|
|
+
|
|
|
+ // Insert into analysisfeatureprop table
|
|
|
$sql = "INSERT INTO {analysisfeatureprop} (analysisfeature_id, type_id, value, rank)".
|
|
|
"VALUES (%d, %d, '%s', %d)";
|
|
|
$previous_db = tripal_db_set_active('chado');
|
|
|
- db_query($sql, $analysisfeature_id, $brite_data_type_id, $mat[2], $hi_rank);
|
|
|
+ db_query($sql, $analysisfeature_id, $brite_data_type_id, $keggline[2], $hi_rank);
|
|
|
tripal_db_set_active($previous_db);
|
|
|
+
|
|
|
+ // Add link to each matched feature
|
|
|
+ if($nid){
|
|
|
+ $matches[2] = preg_replace("/^(.*?)(;\s*\<a)/","<a id=\"tripal_kegg_feature_links\" target=\"_blank\" href=\"".url("/node/$nid")."\">"."$1"."</a>"."$2",$matches[2]);
|
|
|
+ }
|
|
|
}
|
|
|
- // Add link to each matched feature
|
|
|
- $matches[2] = preg_replace("/^(.*?)(;\s*\<a)/","<a id=\"tripal_kegg_feature_links\" target=\"_blank\" href=\"".url("/node/$nid")."\">"."$1"."</a>"."$2",$matches[2]);
|
|
|
- }
|
|
|
- $content .= "<li id=\"term_$id\"><a></a>$matches[2]\n";
|
|
|
+ // we only want to store results with features
|
|
|
+ $ul_content[$level] .= "<li id=\"term_$id\">$matches[2]</li>\n";
|
|
|
+ }
|
|
|
$id++;
|
|
|
}
|
|
|
$content .= "</ul>";
|
spficklin