![](https://investigacion.pucp.edu.pe/grupos/epn/wp-content/uploads/sites/113/2023/03/Figure_1-300x207.jpg)
Toubiana, D.; Maruenda, H. BMC BioInfromatics Journal 22,116:1-14. https://doi.org/10.1186/s12859-021-03994-z
Correlation-based network analysis has become an integral part to study the coordinated behavior of metabolites in plant mapping populations or collections of varieties. To do so pairwise correlations are transformed into network format, where nodes represent the metabolites and the edges the correlations between them. Since metabolites correlation networks aim to follow a small-world topology, some of the edges (correlations) need to be removed. The removal of edges is dependent on two thresholds, i.e. for the correlation coefficient r and its associated p-values. While the selection of the p-value threshold have been well defined, no genuine guidelines have been proposed for the selection of the r-value threshold. In this manuscript we demonstrate how to select the r-value threshold on a metabolite dataset of a potato association panel. We discuss its results in relation to its biological relevance. .
Autor(es):9Toubiana, D.; Maruenda, H. Guidelines for correlation coefficient threshold settings in metabolite correlation networks exemplified on a potato association panel.
Año: 2021
Título de la revista: BMC BioInfromatics Journal
Volumen: 22
Número: 116
Página inicial - Página final: 1-14.
Url: https://doi.org/10.1186/s12859-021-03994-z