A Review of Computational Methods for Clustering Genes with Similar Biological Functions

Abstract

Clustering techniques can group genes based on similarity in biological functions. However, the drawback of using clustering techniques is the inability to identify an optimal number of potential clusters beforehand. Several existing optimization techniques can address the issue. Besides, clustering validation can predict the possible number of potential clusters and hence increase the chances of identifying biologically informative genes. This paper reviews and provides examples of existing methods for clustering genes, optimization of the objective function, and clustering validation. Clustering techniques can be categorized into partitioning, hierarchical, grid-based, and density-based techniques. We also highlight the advantages and the disadvantages of each category. To optimize the objective function, here we introduce the swarm intelligence technique and compare the performances of other methods. Moreover, we discuss the differences of measurements between internal and external criteria to validate a cluster quality. We also investigate the performance of several clustering techniques by applying them on a leukemia dataset. The results show that grid-based clustering techniques provide better classification accuracy; however, partitioning clustering techniques are superior in identifying prognostic markers of leukemia. Therefore, this review suggests combining clustering techniques such as CLIQUE and k-means to yield high-quality gene clusters.

Evaluation of Twenty Genes in Prognosis of Patients with Ovarian Cancer Using Four Different Clustering Methods

“…The previous studies confirmed the proper results of these two methods in the diseases diagnosis and patterning based on gene expression data (Grotkjaer et al, 2006). According to previous studies, k-means is the most well-known method clustering although it suffers from identifying the number of clusters beforehand (Hand and Heard, 2005;Ciaramella and Staiano, 2019;Nies et al, 2019). This fact may lead to poor performance of k-means in gene expression data clustering (Nies et al, 2019).…”

“…Since cancer is a complex disease, it is influenced by various factors, such as upregulation of predisposing factors and downregulation of growth-inhibiting factors which ultimately lead to tumor growth and metastasis. Meanwhile, helpful information about genes or disease pattern could be obtained by comparison of gene expression algorithms under different conditions such as diverse tissues, blood specimens and different growing environments (Hand and Heard, 2005;Ciaramella and Staiano, 2019;Nies et al, 2019). All the genes selected in this study had prominent roles in the control of the activity of the immune system, as well as the chemotaxis, angiogenesis, apoptosis, and so forth.…”

“…According to previous studies, k-means is the most well-known method clustering although it suffers from identifying the number of clusters beforehand (Hand and Heard, 2005;Ciaramella and Staiano, 2019;Nies et al, 2019). This fact may lead to poor performance of k-means in gene expression data clustering (Nies et al, 2019).…”

“…Accordingly, utilizing multivariate statistical analysis methods in examination of gene performance in different diseases has received attention by the researchers interested in this field (Shahraki et al, 2016). Among these, clustering methods are the most important instruments to study the gene microarray data (Grotkjaer et al, 2006;Alavi Majd and Tabatabaei, 2014;Alavimajd et al, 2007;Vahedi et al, 2008;Farhadi and Shahsavani, 2015;Ciaramella and Staiano, 2019;Nies et al, 2019). Clustering can figure out the structure of biomedical data which is helpful for diagnosis or treatment (Nies et al, 2019).…”

Optimal Cluster Head Positioning Algorithm for Wireless Sensor Networks

“…HEED achieves a connected multi-hop inter-cluster network that maximizes the network lifetime for a determined density model and a specified relation between cluster range and transmission range hold. In [ 17 ], Nies et al provided an overview on computational algorithms for genes clustering with correlated biological functions. Different clustering categories have been summarized, where the pros and cons of each category have been discussed.…”