A Comparison of Biclustering with Clustering Algorithms

Singh, Amarjot; Nagrare, Aditya; Srikanth, P.; Kumar, D. M. Vinod; Dwith, Cyn

doi:10.1109/paccs.2011.5990194

Cited by 4 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Biclustering is commonly used in biology especially for clustering gene expression data because traditional clustering methods can't find patterns that are suitable for one cluster. The result from biclustering methods give much better information than clustering methods because of the better enrichment value of the resultant clusters Singh, Nagrare, Srikanth, Kumar, and Dwith (2011). Biclustering can also be interpreted by combining 2 clustering methods.…”

Section: Introductionmentioning

confidence: 99%

Application of BiMax, POLS, and LCM-MBC to Find Bicluster on Interactions Protein between HIV-1 and Human

Bustamam

Siswantining²,

Kaloka³

et al. 2020

AJS

View full text Add to dashboard Cite

Biclustering, in general, is a process of clustering genes and conditions simultaneously rather than clustering them separately. The purpose of biclustering is to discover a subset from experimental data. Further, biclustering results can be analyzed from a biological perspective. Biclustering can also be used for protein-protein interaction. In protein-protein interaction, biclustering can cluster interactions based on rows and columns. In this research, we applied three biclustering algorithms based on graph approach, Binary inclusion-Maximal (BiMax), local search framework based on pairs operation (POLS), and (LCM-MBC) to clustering data of protein-protein interaction between HIV-1 and human. We change the interaction protein-protein interaction data into binary then divided into two datasets called HV positive and HV negative. Then compare the biclustering results of each dataset using heatmap and analyze them with GO terms. From dataset HV positive, BiMax found 30 biclusters, LCM-MBC 31 biclusters, and POLS 13 biclusters. From dataset HV negative, BiMax found eight biclusters, LCM-MBC 14 bicluster, and POLS 10 biclusters. Based on the results of the heatmap, all bicluster entry from BiMax is a protein that interacts, whereas biclusters entry of LCM-MBC and POLS still have proteins that do not interact. It can be concluded that BiMax algorithm is good for clustering protein-protein interaction, especially for binary data.

show abstract

Section: Introductionmentioning

confidence: 99%

Application of BiMax, POLS, and LCM-MBC to Find Bicluster on Interactions Protein between HIV-1 and Human

Bustamam

Siswantining²,

Kaloka³

et al. 2020

AJS

View full text Add to dashboard Cite

show abstract

“…Clustering aims at grouping unlabeled data elements with high similarity into clusters based on any measure obtained solely from the data. These methods have been widely used in different investigative areas such as face detection [10], bioinformatics [9,1,14], market analysis [2] etc. Clustering has been extensively used to detect faces using skin extraction [5] while bioinformatics researchers utilized cluster analysis to build gene groups with related patterns and develop homologous sequences of genes [14].…”

Section: Introductionmentioning

confidence: 99%

Variance Based Moving K-Means Algorithm

Vijay¹,

Raghunath²,

Singh

et al. 2017

2017 IEEE 7th International Advance Computing Conference (IACC)

Self Cite

View full text Add to dashboard Cite

Clustering is a useful data exploratory method with its wide applicability in multiple fields. However, data clustering greatly relies on initialization of cluster centers that can result in large intra-cluster variance and dead centers, therefore leading to sub-optimal solutions. This paper proposes a novel variance based version of the conventional Moving K-Means (MKM) algorithm called Variance Based Moving K-Means (VMKM) that can partition data into optimal homogeneous clusters, irrespective of cluster initialization. The algorithm utilizes a novel distance metric and a unique data element selection criteria to transfer the selected elements between clusters to achieve low intra-cluster variance and subsequently avoid dead centers. Quantitative and qualitative comparison with various clustering techniques is performed on four datasets selected from image processing, bioinformatics, remote sensing and the stock market respectively. An extensive analysis highlights the superior performance of the proposed method over other techniques.

show abstract