dorsal/arxiv
View SchemaMetric learning pairwise kernel for graph inference
| Authors | Jean-Philippe Vert, Jian Qiu, William Stafford Noble |
|---|---|
| Categories | |
| ArXiv ID | q-bio/0610040 |
| URL | https://arxiv.org/abs/q-bio/0610040 |
Abstract
Much recent work in bioinformatics has focused on the inference of various types of biological networks, representing gene regulation, metabolic processes, protein-protein interactions, etc. A common setting involves inferring network edges in a supervised fashion from a set of high-confidence edges, possibly characterized by multiple, heterogeneous data sets (protein sequence, gene expression, etc.). Here, we distinguish between two modes of inference in this setting: direct inference based upon similarities between nodes joined by an edge, and indirect inference based upon similarities between one pair of nodes and another pair of nodes. We propose a supervised approach for the direct case by translating it into a distance metric learning problem. A relaxation of the resulting convex optimization problem leads to the support vector machine (SVM) algorithm with a particular kernel for pairs, which we call the metric learning pairwise kernel (MLPK). We demonstrate, using several real biological networks, that this direct approach often improves upon the state-of-the-art SVM for indirect inference with the tensor product pairwise kernel.
{
"annotation_id": "a1d3ae09-8e83-421c-b2d2-433032605886",
"date_created": "2026-03-02T18:01:35.365000Z",
"date_modified": "2026-03-02T18:01:35.365000Z",
"file_hash": "68c3c75e5106724f81bcbe57ad8827e69bee096ec2866f7b9fc4ce24ada331fc",
"private": false,
"record": {
"abstract": "Much recent work in bioinformatics has focused on the inference of various\ntypes of biological networks, representing gene regulation, metabolic\nprocesses, protein-protein interactions, etc. A common setting involves\ninferring network edges in a supervised fashion from a set of high-confidence\nedges, possibly characterized by multiple, heterogeneous data sets (protein\nsequence, gene expression, etc.). Here, we distinguish between two modes of\ninference in this setting: direct inference based upon similarities between\nnodes joined by an edge, and indirect inference based upon similarities between\none pair of nodes and another pair of nodes. We propose a supervised approach\nfor the direct case by translating it into a distance metric learning problem.\nA relaxation of the resulting convex optimization problem leads to the support\nvector machine (SVM) algorithm with a particular kernel for pairs, which we\ncall the metric learning pairwise kernel (MLPK). We demonstrate, using several\nreal biological networks, that this direct approach often improves upon the\nstate-of-the-art SVM for indirect inference with the tensor product pairwise\nkernel.",
"arxiv_id": "q-bio/0610040",
"authors": [
"Jean-Philippe Vert",
"Jian Qiu",
"William Stafford Noble"
],
"categories": [
"q-bio.QM",
"cs.LG"
],
"title": "Metric learning pairwise kernel for graph inference",
"url": "https://arxiv.org/abs/q-bio/0610040"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "2e5fc636-5e20-4b36-8fd5-c26c2fd6b10a",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}