An investigation on minimum spanning tree based stochastic shuffled multifaceted evolution for efficient clustering of cancer gene data

Abstract

Gene expression analysis has gained more importance in biological newlineand medical research. Traditional methods of studying gene function and newlineregulation were used to focus on one gene at a time. Because many important newlinebiological changes were either missed in an unexpected manner. Over the newlinepast 10 years, numerous methods were developed to allow investigators to newlineexamine mRNA expression levels in thousands of genes in a single newlineexperiment. There is an ample amount of biological data available because of newlineadvanced technologies like Microarray. A microarray measures the newlineexpression of thousands of genes simultaneously which genes are expressed newlineand to what extent. To analyze and retrieve informative knowledge from newlinethese data efficient computational methods are required. The pre-processing newlineof gene expression data should be needed for several reasons. Sometimes, newlinedue to technological problems or mishandling of the microarrays, expression newlinevalues for some genes cannot be measured accurately. This results in the newlinemissing of various data sets. Classification methods do not generally have newlineenough capability to handle missing data. Therefore, the missing values need newlineto be filled in with some reasonable estimates before proceeding with the newlineclassification. newlineWith the rapid advancement of DNA microarray technologies, newlineresearchers in biology can now monitor several gene expression levels in one newlinesingle experiment which is very useful in detecting or classifying specific newlinetypes of tumors. The microarray data and its high-dimensional nature, along newlinewith a small set of samples, have presented a new but well-known newlinephenomenon called the curse of dimensionality. The process of organizing all newlineobjects into various groups based on their traits that are either similar or newlinedissimilar is known as Clustering. newline