array('c', 'GO annotation cutoff', 1000, 'cutoff'),
array('n', 'Number of pairs to fetch', 100, 'number'),
));
foreach ($opts as $k=>$v) {
echo " $k : $v\n";
}
echo "- Testing \biophp\constants\species\spToTaxid\n";
$taxid = \biophp\constants\species::spToTaxid('human');
echo " The taxonomy id of human is: $taxid \n";
echo "- Testing \biophp\constants\genes\idToXref\n";*/
$genes = \biophp\constants\gene::convert(1, 'id', 'ensemblGene');
echo " The ensembl gene id of entrez gene id 1 is: {$genes}\n";
echo " The entrez gene id of ENSG00000183044 is " . \biophp\constants\gene::convert('ENSG00000183044', 'EnsemblGene', 'id') . "\n";
$genes = \biophp\constants\gene::convert(array(1, 10, 100, 101, 102, 103, 104, 107, 108, 109, 111, 112, 113, 115, 116, 117, 118, 119, 12, 120, 123, 124, 126, 127, 128, 13, 130, 131, 132, 134, 135, 136, 14, 140, 141, 142, 143, 146, 148, 15, 150, 151, 152, 153, 154), 'id', 'symbol', true);
//print_r (\biophp\constants\gene::convert($genes, 'synonyms', 'id', true));
print_r($genes);
print_r(\biophp\math\gsea::hg(array("ADORA2A", "ADORA2B", "ADRA1A", "ADRA1D", "ADRB1", "ADRB2", "ADCYAP1R1", "ADORA1", "ADORA3", "ADRA2A", "ADRA2B", "ADRA2C", "ADCY3", "ADCY9", "ADCY1", "ADCY2", "ADCY7", "ADCY6", "ADCY5", "ADA", "ADK", "ADH1A", "ADH1C", "ADH4", "ADH5", "ADH6", "ADH7", "PARP1", "PARP4", "ADAM10", "ADAR", "NAT2", "AANAT", "A1BG", "AADAC", "AAMP", "ADAM8", "ADARB1", "ADCYAP1", "ADD1", "ADD2", "ADD3", "ADPRH", "SERPINA3"), "symbol", "c2.cp.kegg", "symbol", 0.05, 10, false));
print_r(\biophp\math\gsea::ks(array(1, 2, 3, 4, 5), "id", array(4, 5), "id", 1000));
/*
echo "- Testing Naive Bayes\n";
$trainingSet = array(
array(
'data' => array('Outlook'=>'Sunny', 'Temp'=>'Hot', 'Hum'=>'High', 'Windy'=>'Weak'),
'class' => 'No'
),
array(
'data' => array('Outlook'=>'Sunny', 'Temp'=>'Hot', 'Hum'=>'High', 'Windy'=>'Strong'),
'class' => 'No'
),
/**
* @desc: tell whether a given xref id is a kinase
* @param: $xref, the xref id
* @param: $xrefdb, the xref db, same as it in constants.genes
* @param: $species, the species
* Supported species: human, mouse. Default: human
* @return: true|false
*/
public static function isKinaseByXref($xref, $xrefdb, $species = 'human')
{
require_once __DIR__ . "/gene.class.php";
$id = \biophp\constants\gene::xref2Id($xref, $xrefdb, $xrefdb == 'Symbol+Synonyms');
return self::isKinaseById($id, $species);
}
/**
* @desc: do the KS-like gsea without permutation
* @param: $genelist, The sorted gene list
* @param: $informat, The gene id format of the input gene list
* Default: id
* @param: $ref, An array of ref genes
* @param: $refformat, Similar as $informat, the gene format of $ref
* @param: $corr, The correlation of gene in $genelist with the phenotype
* Default: array(1,1,...)
* @param: $weight: The weight of GSEA algorithm
* @return: The enrichment score
*/
public static function es($genelist, $informat = "id", $ref = array(), $refformat = "id", $corr = array(), $weight = 1)
{
if ($informat != "id") {
$genelist = \biophp\constants\gene::convert($genelist, $informat, "id");
}
if ($refformat != "id") {
$ref = \biophp\constants\gene::convert($ref, $refformat, "id");
}
$NR = sizeof(array_intersect($genelist, $ref));
$N = sizeof($genelist);
$Nh = sizeof($ref);
$ref = array_fill_keys($ref, 1);
if (empty($corr)) {
$corr = array_fill(0, $N, 1);
}
$es = -1;
$phit = 0;
$pmiss = 0;
foreach ($genelist as $i => $g) {
if (isset($ref[$g])) {
$phit += pow($corr[$i], $weight) / $NR;
} else {
$pmiss += 1 / ($N - $Nh);
}
if ($phit - $pmiss > $es) {
$es = $phit - $pmiss;
}
}
return $es;
}