Guidelines for correlation coefficient threshold settings in metabolite correlation networks exemplified on a potato association panel

Abstract

Background Correlation network analysis has become an integral tool to study metabolite datasets. Networks are constructed by omitting correlations between metabolites based on two thresholds—namely the r and the associated p-values. While p-value threshold settings follow the rules of multiple hypotheses testing correction, guidelines for r-value threshold settings have not been defined. Results Here, we introduce a method that allows determining the r-value threshold based on an iterative approach, where different networks are constructed and their network topology is monitored. Once the network topology changes significantly, the threshold is set to the corresponding correlation coefficient value. The approach was exemplified on: (i) a metabolite and morphological trait dataset from a potato association panel, which was grown under normal irrigation and water recovery conditions; and validated (ii) on a metabolite dataset of hearts of fed and fasted mice. For the potato normal irrigation correlation network a threshold of Pearson’s |r|≥ 0.23 was suggested, while for the water recovery correlation network a threshold of Pearson’s |r|≥ 0.41 was estimated. For both mice networks the threshold was calculated with Pearson’s |r|≥ 0.84. Conclusions Our analysis corrected the previously stated Pearson’s correlation coefficient threshold from 0.4 to 0.41 in the water recovery network and from 0.4 to 0.23 for the normal irrigation network. Furthermore, the proposed method suggested a correlation threshold of 0.84 for both mice networks rather than a threshold of 0.7 as applied earlier. We demonstrate that the proposed approach is a valuable tool for constructing biological meaningful networks.

NMF-Based Approach for Missing Values Imputation of Mass Spectrometry Metabolomics Data

“…Threshold restrictions were applied to the correlation coefficients and their associated p -values to identify spurious correlations between metabolites. The non-significant edges were removed from CCN to ensure the sparsity of the network [ 21 ]. As a result, the sparsity of the original networks of datasets are as follows: 0.182 (Dataset I), 0.175 (Dataset II), 0.153 (Dataset III), and 0.127 (Dataset IV).…”

Correlation-based network analysis combined with machine learning techniques highlights rosmarinic acid biosynthesis activation in tuber of potato plants corresponding to water recovery treatment

“…To analyze the differences of coordinated behavior of metabolites between the two treatments, we employed correlation-based network analysis, where the nodes in the network correspond to the metabolites and edges between them the computed correlation coefficient CNs (Toubiana and Maruenda 2021). By estimating network properties and using set theory applied to networks, we highlighted the differences between the PS and NS networks.…”

“…In the CN, the nodes represent compounds and the edges between them the significant correlation coefficients (r). Thresholds for both networks were set to r ≥ 0.6 and a p-value of  0.01 to remove spurious correlations as determined using guidelines to detect r-and p-value thresholds in CNs (Toubiana and Maruenda 2021) (Supplementary Figure S1). Both CNs were composed of 685 nodes, corresponding to the compounds in the dataset.…”

Construction of a PANoptosis-Related LncRNA Prognostic Model and Study on the Function of Linc01135 for Gastric Cancer

“…To pinpoint potential LncRNAs linked to PANoptosis [25] , we utilized the limma package in R software to conduct a co-expression correlation analysis between the expression profiles of LncRNA and PANoptosis genes. To ensure significance, we applied stringent screening criteria, considering only those with a correlation coefficient of |cor| > 0.4 and a statistical significance level of P < 0.001 [26].…”