Methods for genomics and trascriptomics analysis

The recent technology development for large-scale sequencing, for the analysis of gene expression using microarray-based technologies, for large-scale point mutations and SNP analysis, has revealed some serious limitations in the management and analysis of the data generated.

The CNR-Bioinformatics project is likely to extend CNR's collaboration with other initiatives promoted by MIUR, both in FIRB projects (LITBIO, LIBI) and FAR projects, recently approved with the aim of achieving a synergy between these projects. In particular, this platform will permit the analysis of genes, promoter regions, genetic expression, EST, Single Nucleotide Polymorphisms SNP, mutation research and metabolic network studies.

Furthermore statistic models and algorithms will be developed, with experimental validation, for genomic and trascriptomics data analysis aimed at studying principal biological processes and at the comprehension of genic expression regulation mechanisms. The methods will be developed with the statistical learning theory and will be applied to the analysis of data provided by gene expression microarray experiments and analysis of SNP.

Here new innovative methods will be developed, based on mathematics and informatics advanced technologies for the comparison of genomes and the phylogenetic analysis between different species, and appropriate intelligent methods for the extraction of information in genomics and proteomics sequences.

One of the long-term objectives of modern biology is the comprehension of cell regulation mechanisms. This objective requires, in a preliminary step, the comprehension of gene regulation mechanisms in specific and simplified contests. To this end the analysis of microrray data seems to be very useful both for the identification of the genes that play an important role in diseases of genetic origin, and for the identification of their transcription sites.

In recent literature different statistic methods for the analysis of genomic data orientated to 'classic' experiments analysis have been proposed. However, in experiments whose purpose is to monitor the change in profile of gene expression against time (for example the study of the genetic-temporal effect in a cell's drug treatment), the availability of statistically reliable and/or efficient methods within the scientific community is poor. The project has therefore scheduled a research activity in this reseach area.

Furthermore statistic algorithms for genes and/or data expression clustering profiles from microarray will be developed. Later on, the focus will move to the study of statistical techniques for the indentification of co-regulated transcription sites. The knowledge of these sites is fundamental for the comprehension of the regulation mechanisms of these genes.