A Systematic Comparison of Genome Scale Clustering Algorithms

Abstract

Background: A wealth of clustering algorithms has been applied to gene co-expression experiments. These algorithms cover a broad range of approaches, from conventional techniques such as k-means and hierarchical clustering, to graphical approaches such as k-clique communities, weighted gene co-expression networks (WGCNA) and paraclique. Comparison of these methods to evaluate their relative effectiveness provides guidance to algorithm selection, development and implementation. Most prior work on comparative clustering evaluation has focused on parametric methods. Graph theoretical methods are recent additions to the tool set for the global analysis and decomposition of microarray co-expression matrices that have not generally been included in earlier methodological comparisons. In the present study, a variety of parametric and graph theoretical clustering algorithms are compared using well-characterized transcriptomic data at a genome scale from Saccharomyces cerevisiae. Methods: For each clustering method under study, a variety of parameters were tested. Jaccard similarity was used to measure each cluster's agreement with every GO and KEGG annotation set, and the highest Jaccard score was assigned to the cluster. Clusters were grouped into small, medium, and large bins, and the Jaccard score of the top five scoring clusters in each bin were averaged and reported as the best average top 5 (BAT5) score for the particular method. Results: Clusters produced by each method were evaluated based upon the positive match to known pathways. This produces a readily interpretable ranking of the relative effectiveness of clustering on the genes. Methods were also tested to determine whether they were able to identify clusters consistent with those identified by other clustering methods.

Ontogenetic changes in embryonic and brain gene expression in progeny produced from migratory and residentOncorhynchus mykiss

“…This range includes both cluster‐based methods which are executed in a manner independent of known interactions and pathways for genes, evaluating correlation or co‐expression of genes in a study (Jay et al . ), and methods that evaluate how specific biological functions or pathways are ‘enriched’ for differentially expressed genes (Khatri et al . ; Mitrea et al .…”

“…Finally, there are many methods available to explore the integration and functional significance of gene networks that are differentially regulated. This range includes both cluster-based methods which are executed in a manner independent of known interactions and pathways for genes, evaluating correlation or co-expression of genes in a study (Jay et al 2012), and methods that evaluate how specific biological functions or pathways are 'enriched' for differentially expressed genes (Khatri et al 2012;Mitrea et al 2013). We took the latter approach, in an attempt to evaluate a priori pathways and functions and how those particular pathways and functions are differentially regulated over development, with the caveat that such pathways are built largely on function in mammalian systems.…”

Beyond modules and hubs: the potential of gene coexpression networks for investigating molecular mechanisms of complex brain disorders

“…A plethora of methods can identify putative coexpression modules (Fortunato ; Jay et al ; Langfelder & Horvath ). Choosing the ‘best’ clustering method is a balance between the mathematical ability to detect locally dense modules, the biological ability to find functionally enriched clusters and computational efficiency.…”

Cancer Stem Cell Gene Profile as Predictor of Relapse in High Risk Stage II and Stage III, Radically Resected Colon Cancer Patients

“…To evaluate different patterns of gene expression, we conducted clustering analysis by K-means (K = 2) method [15]. This particular clustering analysis algorithm was chosen to test the possible existence of 2 different patterns of gene expression, related to different prognosis.…”