Francesco Masulli

Biclustering Bioinformatics Data Sets: A Possibilistic Approach

The analysis of genomic data from DNA microarray can produce a valuable information on the biological relevance of genes and correlations among them. In the last few years some biclustering techniques have been proposed and applied to this analysis.


Biclustering is a learning task for finding clusters of samples possessing similar characteristics together with features creating these similarities. When applied to genomic data it can allow us to identify genes with similar behavior with respect to different conditions. In this talk a new approach to the biclustering problem will be introduced extending the Possibilistic Clustering paradigm.

The proposed Possibilistic Biclustering algorithm finds one bicluster at a time, assigning a membership to the bicluster for each gene and for each condition. Some results on oligonucleotide microarray data sets will be presented and compared with those obtained using other biclustering methods.