|
@@ -141,6 +141,7 @@ function get_BLAST_job($job_id) {
|
|
$job->files->result->xml = $blastjob->result_filestub . '.xml';
|
|
$job->files->result->xml = $blastjob->result_filestub . '.xml';
|
|
$job->files->result->tsv = $blastjob->result_filestub . '.tsv';
|
|
$job->files->result->tsv = $blastjob->result_filestub . '.tsv';
|
|
$job->files->result->html = $blastjob->result_filestub . '.html';
|
|
$job->files->result->html = $blastjob->result_filestub . '.html';
|
|
|
|
+ $job->files->result->gff = $blastjob->result_filestub . '.gff';
|
|
|
|
|
|
return $job;
|
|
return $job;
|
|
}
|
|
}
|
|
@@ -167,6 +168,7 @@ function run_BLAST_tripal_job($program, $query, $database, $output_filestub, $op
|
|
$output_file_xml = $output_filestub . '.xml';
|
|
$output_file_xml = $output_filestub . '.xml';
|
|
$output_file_tsv = $output_filestub . '.tsv';
|
|
$output_file_tsv = $output_filestub . '.tsv';
|
|
$output_file_html = $output_filestub . '.html';
|
|
$output_file_html = $output_filestub . '.html';
|
|
|
|
+ $output_file_gff = $output_filestub . '.gff';
|
|
|
|
|
|
print "\nExecuting $program\n\n";
|
|
print "\nExecuting $program\n\n";
|
|
print "Query: $query\n";
|
|
print "Query: $query\n";
|
|
@@ -243,6 +245,19 @@ function run_BLAST_tripal_job($program, $query, $database, $output_filestub, $op
|
|
);
|
|
);
|
|
}
|
|
}
|
|
|
|
|
|
|
|
+ print "\tGFF\n";
|
|
|
|
+ convert_tsv2gff3($output_file_tsv,$output_file_gff);
|
|
|
|
+
|
|
|
|
+ if (!file_exists($output_file_gff)) {
|
|
|
|
+ tripal_report_error(
|
|
|
|
+ 'blast_ui',
|
|
|
|
+ TRIPAL_WARNING,
|
|
|
|
+ "Unable to convert BLAST Tabular Output to GFF Output (%archive => %file).",
|
|
|
|
+ array('%archive' => $output_file, '%file' => $output_file_gff),
|
|
|
|
+ array('print' => TRUE)
|
|
|
|
+ );
|
|
|
|
+ }
|
|
|
|
+
|
|
print "\tHTML (includes alignments)\n";
|
|
print "\tHTML (includes alignments)\n";
|
|
system("$blast_formatter_command -archive $output_file -outfmt 0 -out $output_file_html -html");
|
|
system("$blast_formatter_command -archive $output_file -outfmt 0 -out $output_file_html -html");
|
|
if (!file_exists($output_file_tsv)) {
|
|
if (!file_exists($output_file_tsv)) {
|
|
@@ -668,4 +683,94 @@ function generate_blast_hit_image($acc = '', $scores, $hits, $tsize, $qsize, $na
|
|
ob_end_clean();
|
|
ob_end_clean();
|
|
|
|
|
|
return $b64_img;
|
|
return $b64_img;
|
|
-}
|
|
|
|
|
|
+}
|
|
|
|
+
|
|
|
|
+/** convert tsv blast output to gff output file
|
|
|
|
+ *
|
|
|
|
+ * Created by Sofia Robb
|
|
|
|
+ * 09/15/2016
|
|
|
|
+ *
|
|
|
|
+ * the subject (hit) will be the source feature
|
|
|
|
+ * the query will be the target
|
|
|
|
+ *
|
|
|
|
+ * @param $blast_tsv, $blast_gff
|
|
|
|
+ * the name of the blast tsv output file
|
|
|
|
+ * the name of the blast gff output file
|
|
|
|
+ *
|
|
|
|
+ * @return
|
|
|
|
+ * returns nothing but creates the gff file in the same dir as the tsv with gff extension
|
|
|
|
+ */
|
|
|
|
+
|
|
|
|
+function convert_tsv2gff3($blast_tsv,$blast_gff){
|
|
|
|
+ $gff = fopen($blast_gff,"w");
|
|
|
|
+ fwrite($gff,"##gff-version 3\n");
|
|
|
|
+
|
|
|
|
+ $results = array();
|
|
|
|
+
|
|
|
|
+ $tsv = fopen($blast_tsv, "r") or die("Unable to open tsv file!");
|
|
|
|
+ while(!feof($tsv)) {
|
|
|
|
+ $line = fgets($tsv);
|
|
|
|
+ $line = rtrim($line);
|
|
|
|
+ if (preg_match('/^#/',$line) or preg_match('/^\s*$/',$line)){
|
|
|
|
+ continue;
|
|
|
|
+ }
|
|
|
|
+ //$line has these parts: $queryId, $subjectId, $percIdentity, $alnLength, $mismatchCount,$gapOpenCount, $queryStart, $queryEnd, $subjectStart,$subjectEnd, $eVal, $bitScore
|
|
|
|
+ $parts = preg_split('/\t/', $line);
|
|
|
|
+ $hitname = $parts[1];
|
|
|
|
+ $hspInfo = "$parts[0],$parts[8],$parts[9],$parts[6],$parts[7]";
|
|
|
|
+ $eval = $parts[10];
|
|
|
|
+ $results[$hitname][$hspInfo]=$eval;
|
|
|
|
+ $IDs = array();
|
|
|
|
+ $count;
|
|
|
|
+ $last_q;
|
|
|
|
+ //need to go thru each line of tsv to find the first and last hsp of a hit.
|
|
|
|
+ //need to get the smallest and largest coordinate for the parent feature
|
|
|
|
+ foreach ($results as $s => $hspInfoArray) {
|
|
|
|
+ $hsp = 0;
|
|
|
|
+ foreach ($hspInfoArray as $hspInfoStr => $e){
|
|
|
|
+ list($q,$ss,$se,$qs,$qe) = preg_split('/,/',$hspInfoStr);
|
|
|
|
+ if (!$last_q or $q != $last_q){
|
|
|
|
+ $count++;
|
|
|
|
+ $hsp=0;
|
|
|
|
+ }
|
|
|
|
+ $q_strand = '+';
|
|
|
|
+ if ($qs > $qe){
|
|
|
|
+ list($qs,$qe) = array($qe,$qs);
|
|
|
|
+ $q_strand = '-';
|
|
|
|
+ }
|
|
|
|
+ $IDs["$s,$q"]['strand']='+';
|
|
|
|
+ list($start,$end) = array($ss,$se);
|
|
|
|
+ if($ss > $se){
|
|
|
|
+ list($start,$end) = array($se,$ss);
|
|
|
|
+ $IDs["$s,$q"]['strand']='-';
|
|
|
|
+ }
|
|
|
|
+ if (!array_key_exists('SS',$IDs["$s,$q"]) or $ss < $IDs["$s,$q"]['SS']){
|
|
|
|
+ $IDs["$s,$q"]['SS'] = $ss;
|
|
|
|
+ }
|
|
|
|
+ if (!array_key_exists('SE',$IDs["$s,$q"]) or $se > $IDs["$s,$q"]['SE']){
|
|
|
|
+ $IDs["$s,$q"]['SE'] = $se;
|
|
|
|
+ }
|
|
|
|
+ if (!array_key_exists('E',$IDs["$s,$q"]) or $e < $IDs["$s,$q"]['E']){
|
|
|
|
+ $IDs["$s,$q"]['E'] = $e;
|
|
|
|
+ }
|
|
|
|
+ $hsp++;
|
|
|
|
+ $IDs["$s,$q"]['HSPs'][] = join("\t", array($s, "BLASTRESULT" , "match_part" , $start , $end , $e , $IDs["$s,$q"]['strand'] , '.' , "ID=$s.$count.$hsp;Parent=$s.$count;Target=$q $qs $qe $q_strand"));
|
|
|
|
+ $last_q = $q;
|
|
|
|
+
|
|
|
|
+ }
|
|
|
|
+ }
|
|
|
|
+
|
|
|
|
+ }//end of tsv file while
|
|
|
|
+ // now can print a parent gff line and all the children
|
|
|
|
+ // the evalues seem to be sorted properly without actually sorting them, need to make sure this is always true
|
|
|
|
+ foreach ($IDs as $sq => $value ){
|
|
|
|
+ list ($s,$q) = preg_split('/,/' , $sq);
|
|
|
|
+ $evalue = $IDs["$s,$q"]['E'];
|
|
|
|
+ $parent = join ("\t", array($s, "BLASTRESULT" , "match" , $IDs["$s,$q"]['SS'] , $IDs["$s,$q"]['SE'] , $IDs["$s,$q"]['E'] , $IDs["$s,$q"]['strand'] , '.' , "ID=$s.$count;Name=$q($evalue)")) . "\n";
|
|
|
|
+ $child = join ("\n",$IDs[$sq]['HSPs']) . "\n";
|
|
|
|
+ fwrite($gff,$parent);
|
|
|
|
+ fwrite($gff,$child);
|
|
|
|
+ }
|
|
|
|
+ fclose($tsv);
|
|
|
|
+ fclose($gff);
|
|
|
|
+}
|