Relational clustering for gene expression profiles and drug activity pattern analysis

The combined analysis of the micro array and drug-activity datasets has the potential of revealing valuable knowledge about various relations among gene expressions and drug activity patterns in malignant cells. However, the huge amount of biological data needs appropriate data mining models in order to extract interesting patterns and useful information. The ultimate goal of the paper is to define a model which, given the gene expression profile related to a specific tumor tissue, could help in selecting a set of most responsive drugs. This is accomplished through an unsupervised classification algorithm that associates to a cell line the set of drugs that most probably are related to its gene expression profile. The classification engine is based on a Relational Clustering algorithm which groups cell lines using drug response information and taking into account cell-to-cell relationships defined by the similarity of their gene expression profiles.