A Review on clustering and visualization methodologies for Genomic data analysis

This abstract presents a survey on the aims, the problems and the meth-ods concerning Cluster Analysis and its applications in genomic data analysis. With the term Cluster Analysis we refer to a data exploration tool whose goal is grouping objects of similar kind into their respective categories without a priori information on their classes. We can look at cluster analysis as a classification problem with no labeled samples, or without any a priori knowledge about the way the objects have to be put together. There are several and heterogeneous problems linked to the cluster analysis and several times they are treated sepa-rately. In this work we examine these problems, and we illustrate the different approaches and their applications to Computational Biology and Bioinformat-ics. The problems related to Cluster Analysis in the context of high-dimensional genomic data analysis can be summarized as shown in figure 1. In this figure each node represents an item of the data exploration problem via cluster analysis or computational methods used in this kind of data analysis. The edges of this graph can be mono-directional or bi-directional and, following a path (accordin