A SURVEY ON THE CLUSTERING METHODS FOR THE IDENTIFICATION OF COEXPRESSED GENES

Electronics and Communication Engineering

NATIONAL CONFERENCE ON TECHNOLOGICAL ADVANCEMENTS IN ENGINEERING

Bioinformatics is an interdisciplinary area of collecting and analyzing gene codes .Gene clustering is one of the first steps in gene expression analysis. The main aim of clustering is to subdivide a set of data in such a way that similar data fall into same cluster whereas dissimilar items fall in different clusters. The main application is Gene-expression profiling, medical diagnosis – cancers, cardio vascular disorders, genetic disorders, biomedicine, drug discovery and development and tumor classification. This paper is the comparative study of different types of clustering techniques like k-means, fuzzy c means, self organizing map, hierarchical clustering. For the precise clustering of genes it is important to choose optimal number of genes and feature selection from microarray dataset. So dimensionality reduction is necessary in gene clustering. The gene expression data is mainly analyzed in two ways, supervised and unsupervised. Clustering techniques have proven to be helpful to understand gene function, cellular processes, and subtypes of cells. The different methods of measuring similarity are by considering Euclidean distance, pearson correlation and Manhattan distance etc .There is no one-size-fits all solution to clustering. It is ambiguous of how good clusters look like. So validity of clusters should be done in the sense of biological significance.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.