Correlation-guided Network Integration (CoNI), an R package for integrating numerical omics data that allows multiform graph representations to study molecular interaction networks

Bioinform Adv. 2022 Jun 6;2(1):vbac042. doi: 10.1093/bioadv/vbac042. eCollection 2022.

Authors

José Manuel Monroy Kuhn^{1

2}, Viktorian Miok^{1

2

3}, Dominik Lutter^{1

2}

Affiliations

¹ Computational Discovery Unit, Institute for Diabetes & Obesity, Helmholtz Zentrum München, Neuherberg, Germany.
² German Center for Diabetes Research (DZD), Neuherberg, Germany.
³ Astrocyte-Neuron Networks, Institute for Diabetes & Obesity, Helmholtz Zentrum München, Neuherberg, Germany.

Abstract

Summary: Today's immense growth in complex biological data demands effective and flexible tools for integration, analysis and extraction of valuable insights. Here, we present CoNI, a practical R package for the unsupervised integration of numerical omics datasets. Our tool is based on partial correlations to identify putative confounding variables for a set of paired dependent variables. CoNI combines two omics datasets in an integrated, complex hypergraph-like network, represented as a weighted undirected graph, a bipartite graph, or a hypergraph structure. These network representations form a basis for multiple further analyses, such as identifying priority candidates of biological importance or comparing network structures dependent on different conditions.

Availability and implementation: The R package CoNI is available on the Comprehensive R Archive Network (https://cran.r-project.org/web/packages/CoNI/) and GitLab (https://gitlab.com/computational-discovery-research/coni). It is distributed under the GNU General Public License (version 3).

Supplementary information: Supplementary data are available at Bioinformatics Advances online.