A Novel Method for the Analysis of Gene Expression Microarray Data with K-Means Clustering: Sorted K-Means

Background: Microarray technology has revolutionized the way genomic analysis has been performed. High-throughput data acquisition, brought up a challenge in data comprehension i.e. in gene expression. Methods: k-means cluster obtained after analysis of miRNA expression data have been sorted by an algorithmic procedure. Results: The proposed method managed to sort k-means centroids and manifest a more simple way of drawing conclusions on studied tumor samples. miRNAs were unraveled that changed in expression levels with respect to tumor aggressiveness. Conclusions: In the present work we presented a new and simple approach in data analysis using a new analysis approach, which we termed sorted-k-means analysis. Keywords—k-means, microarrays, sorted k-means, gene expression, miRNA, tumors.