dorsal/arxiv
View SchemaProbabilistic Clustering of Sequences: Inferring new bacterial regulons by comparative genomics
| Authors | Erik van Nimwegen, Mihaela Zavolan, Nikolaus Rajewsky, Eric D. Siggia |
|---|---|
| Categories | |
| ArXiv ID | physics/0206045 |
| URL | https://arxiv.org/abs/physics/0206045 |
| DOI | 10.1073/pnas.112690399 |
| Journal | PNAS 99 (2002) 7323-7328 (This ArXiv version is slighly modified and contains the "suppporting text" as appendices.) |
Abstract
Genome wide comparisons between enteric bacteria yield large sets of conserved putative regulatory sites on a gene by gene basis that need to be clustered into regulons. Using the assumption that regulatory sites can be represented as samples from weight matrices we derive a unique probability distribution for assignments of sites into clusters. Our algorithm, 'PROCSE' (probabilistic clustering of sequences), uses Monte-Carlo sampling of this distribution to partition and align thousands of short DNA sequences into clusters. The algorithm internally determines the number of clusters from the data, and assigns significance to the resulting clusters. We place theoretical limits on the ability of any algorithm to correctly cluster sequences drawn from weight matrices (WMs) when these WMs are unknown. Our analysis suggests that the set of all putative sites for a single genome (e.g. E. coli) is largely inadequate for clustering. When sites from different genomes are combined and all the homologous sites from the various species are used as a block, clustering becomes feasible. We predict 50-100 new regulons as well as many new members of existing regulons, potentially doubling the number of known regulatory sites in E. coli.
{
"annotation_id": "72ea3e9e-8f09-4bd1-ae05-17647b3bcb0a",
"date_created": "2026-03-02T18:00:39.479000Z",
"date_modified": "2026-03-02T18:00:39.479000Z",
"file_hash": "a6dfbd2852590234cb3da64f0d964e3e230913d82b57698544ef61661d9f7b77",
"private": false,
"record": {
"abstract": "Genome wide comparisons between enteric bacteria yield large sets of\nconserved putative regulatory sites on a gene by gene basis that need to be\nclustered into regulons. Using the assumption that regulatory sites can be\nrepresented as samples from weight matrices we derive a unique probability\ndistribution for assignments of sites into clusters. Our algorithm, \u0027PROCSE\u0027\n(probabilistic clustering of sequences), uses Monte-Carlo sampling of this\ndistribution to partition and align thousands of short DNA sequences into\nclusters. The algorithm internally determines the number of clusters from the\ndata, and assigns significance to the resulting clusters. We place theoretical\nlimits on the ability of any algorithm to correctly cluster sequences drawn\nfrom weight matrices (WMs) when these WMs are unknown. Our analysis suggests\nthat the set of all putative sites for a single genome (e.g. E. coli) is\nlargely inadequate for clustering. When sites from different genomes are\ncombined and all the homologous sites from the various species are used as a\nblock, clustering becomes feasible. We predict 50-100 new regulons as well as\nmany new members of existing regulons, potentially doubling the number of known\nregulatory sites in E. coli.",
"arxiv_id": "physics/0206045",
"authors": [
"Erik van Nimwegen",
"Mihaela Zavolan",
"Nikolaus Rajewsky",
"Eric D. Siggia"
],
"categories": [
"physics.bio-ph",
"physics.data-an",
"q-bio.GN",
"q-bio.QM"
],
"doi": "10.1073/pnas.112690399",
"journal_ref": "PNAS 99 (2002) 7323-7328 (This ArXiv version is slighly modified\n and contains the \"suppporting text\" as appendices.)",
"title": "Probabilistic Clustering of Sequences: Inferring new bacterial regulons by comparative genomics",
"url": "https://arxiv.org/abs/physics/0206045"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "0c9c481c-af84-4f0e-91f1-6089fd2f30f7",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}