dorsal/arxiv
View SchemaHighProbability determines which alternative hypotheses are sufficiently probable: Genomic applications include detection of differential gene expression
| Authors | David R. Bickel |
|---|---|
| Categories | |
| ArXiv ID | q-bio/0402049 |
| URL | https://arxiv.org/abs/q-bio/0402049 |
Abstract
Many genomic experiments, notably microarray experiments seeking to detect differential gene expression, involve calculating a large number of p-values. This leads to the multiple testing problem: when the number of null hypotheses is large, the probability of accepting at least one false alternative hypothesis is often much greater than the significance level of the tests, which tends to mislead investigators. Software called HighProbability provides a simple, fast, reliable solution to the multiple testing problem, with applications to many areas of bioinformatics. For example, in a microarray study, HighProbability can determine which genes are probably differentially expressed. Given a set of p-values not adjusted for multiple testing, HighProbability determines which ones are low enough to imply a high probability of the truth of their alternative hypotheses. The set of p-values may be determined by conventional hypothesis testing or by random permutations using existing R or S-PLUS software. HighProbability is freely available under license through http://www.davidbickel.com . Coded in S, HighProbability currently requires an installation of R or S-PLUS, but the algorithm is short enough for fast implementation in non-S languages as well.
{
"annotation_id": "c6414de9-a0e8-48c6-884b-42e77e83b13d",
"date_created": "2026-03-02T18:01:31.419000Z",
"date_modified": "2026-03-02T18:01:31.419000Z",
"file_hash": "78be0379e6c63e8c8933508b77680422eed3e2b78679403d58102ffb52ef24f8",
"private": false,
"record": {
"abstract": "Many genomic experiments, notably microarray experiments seeking to detect\ndifferential gene expression, involve calculating a large number of p-values.\nThis leads to the multiple testing problem: when the number of null hypotheses\nis large, the probability of accepting at least one false alternative\nhypothesis is often much greater than the significance level of the tests,\nwhich tends to mislead investigators. Software called HighProbability provides\na simple, fast, reliable solution to the multiple testing problem, with\napplications to many areas of bioinformatics. For example, in a microarray\nstudy, HighProbability can determine which genes are probably differentially\nexpressed. Given a set of p-values not adjusted for multiple testing,\nHighProbability determines which ones are low enough to imply a high\nprobability of the truth of their alternative hypotheses. The set of p-values\nmay be determined by conventional hypothesis testing or by random permutations\nusing existing R or S-PLUS software. HighProbability is freely available under\nlicense through http://www.davidbickel.com . Coded in S, HighProbability\ncurrently requires an installation of R or S-PLUS, but the algorithm is short\nenough for fast implementation in non-S languages as well.",
"arxiv_id": "q-bio/0402049",
"authors": [
"David R. Bickel"
],
"categories": [
"q-bio.QM",
"q-bio.MN"
],
"title": "HighProbability determines which alternative hypotheses are sufficiently probable: Genomic applications include detection of differential gene expression",
"url": "https://arxiv.org/abs/q-bio/0402049"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "20b5bf32-115c-4070-9c2f-0dd7a99e7d28",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}