An additional k-means clustering step improves the biological features of WGCNA gene co-expression networks

Abstract

BackgroundWeighted Gene Co-expression Network Analysis (WGCNA) is a widely used R software package for the generation of gene co-expression networks (GCN). WGCNA generates both a GCN and a derived partitioning of clusters of genes (modules). We propose k-means clustering as an additional processing step to conventional WGCNA, which we have implemented in the R package km2gcn (k-means to gene co-expression network, https://github.com/juanbot/km2gcn).ResultsWe assessed our method on networks created from UKBEC data (10 different human brain tissues), on networks created from GTEx data (42 human tissues, including 13 brain tissues), and on simulated networks derived from GTEx data. We observed substantially improved module properties, including: (1) few or zero misplaced genes; (2) increased counts of replicable clusters in alternate tissues (x3.1 on average); (3) improved enrichment of Gene Ontology terms (seen in 48/52 GCNs) (4) improved cell type enrichment signals (seen in 21/23 brain GCNs); and (5) more accurate partitions in simulated data according to a range of similarity indices.ConclusionsThe results obtained from our investigations indicate that our k-means method, applied as an adjunct to standard WGCNA, results in better network partitions. These improved partitions enable more fruitful downstream analyses, as gene modules are more biologically meaningful.Electronic supplementary materialThe online version of this article (doi:10.1186/s12918-017-0420-6) contains supplementary material, which is available to authorized users.

Identification of three m6A‐related mRNAs signature and risk score for the prognostication of hepatocellular carcinoma

“…The expression data profile of the selected genes was qualified, and the samples were clustered to detect outliers. Gene clustering modules were identified based on the clinical features (including the expression of the m6A regulatory genes that we selected before) and topological overlap matrix‐based dissimilarity . Then, the correlations between module eigengenes and clinical traits were calculated to identify the relevant modules.…”

An integrated genomic approach to dissect the genetic landscape regulating the cell-to-cell transfer of a-synuclein

“…These residuals, along with the networks and annotations are accessible, for each tissue at the CoExpGTEx GitHub repository https://github.com/juanbot/CoExpGTEx. The coexpression networks are obtained with the WGCNA R package (Langfelder and Horvath 2008) and an additional refinement step of the clusters, based on the k-means algorith implemented in the CoExpNets R package (Botia et al, 2017). We constructed a set of clusters from gene expression values based on correlation between genes across samples through building a gene expression adjacency matrix (with scale free topology).…”

“…WGCNA was performed using the GTEx version 6 dataset. The modules within each tissue assayed were generated using the recently developed method with an additional k-means clustering step (Botia et al, 2017). We first assessed the ability of WGCNA to correctly identify functional relationships between genes and to correctly annotate those relationships and their enrichment for particular cell type markers.…”

G2P: Using machine learning to understand and predict genes causing rare neurological disorders

“…Using transcriptomic data generated by the GTEx 25 project and covering 47 human tissues (including 13 brain regions, tibial nerve and skeletal muscle amongst others) we generated measures of tissue-specific gene-expression and co-expression (Online Methods). For each of the 47 human tissues we created a co-expression network using Weighted Gene Coexpression Network Analysis 17 optimized by k-means 18 . This provided estimates of each gene's global and local connectivity in relation to all other expressed genes in the tissue (as captured by the terms "adjacency" and "module membership", Online Methods).…”