Motivation & Background

Genetics is a science that has undergone great evolution in the last fifty years, starting with the discovery of DNA, and is still promising a great expansion to new discoveries that may help in understanding, for example, the treatment and prevention of many diseases. A very important role in this direction can be played by mathematics and statistics, with their ability to produce models and methods to explain/analyze data and the phenomena in question, and by informatics that can provide the hardware and software tools indispensable to manipulate the great mass of data involved.

In fact, enormous amounts of data from experimental protocols and platforms are currently available in public databases; the joint use of such data and the subsequent analysis with specific methodologies can lead to new discoveries without requiring additional experimental costs. Furthermore, analysis of experiments will require new increasingly sophisticated and accurate analytical tools in order to take into account the complexity of the phenomena in question, in which many variables are involved. This leads to a natural involvement in Science Informatics, called to provide the hardware and software tools to perform a quick and careful analysis, using very powerful computers (HPC) or based on GRID computing platforms.

In particular, Bioinformatics has allowed the study of mathematical models and algorithms for the representation of special problems that arise in the study of DNA (genomics) and proteins (proteomics). Some of the most important research that will be addressed includes: functional characterization of genes, phylogeny and phylogenetic trees, DNA recombination, problems of multiple alignment of DNA and protein sequences, prediction of the three-dimensional structure of proteins, study of gene networks, study of regulatory processes, interaction between proteins, analysis and interpretation of gene expression data, research for polymorphisms, SNP and mutations, correlation between genetic diseases and clinical diagnosis.

The huge flow of data sequences derived from projects sequencing of entire genomes of different organisms at various levels of evolution has introduced the problem of analysing the structures and functions of genes. Informatics systems are increasingly required in order to cope with the processing of data in molecular biology and medicine. The complexity of information, along with the considerable amount of data available, their continuous update, and the difficulty of use of many programs of analysis useful for investigations in the biomedical field, makes an everyday use of these investigation tools, which have now become essential in many industrial and educational sectors, problematic.

The introduction of high-performance equipment used in DNA sequencing and the continuous technological progress in studies of gene expression, as well as in studies of population and proteomics, has highlighted the absence in Italy of an adequate system for managing Data and analysis aimed at complex systems and their models designed in System Biology, Bioinformatics and Medical Informatics.

The most recent techniques for the study of gene expression, first used only for research purposes, were widely available in hospitals, analysis laboratories and more generally as routine tools in identifying genetic diseases. Using these technologies requires the presence of new professional figures, specific to Bioinformatics.

In the last few years, Many American and Anglo-Saxon universities have departments which specialize in biostatistics, bioinformatics, bioengineering, biophysics skills in order to facilitate studies and applications in genetics and medicine. Various specific degrees and doctorates for these professions have also been created. In this context biologists, geneticists, doctors, physicists, mathematical chemists, informaticians and statisticians all work together.

In Italy the current situation presents a significant delay. In fact, opportunities for research in this new interdisciplinary field, which is currently in full worldwide development, do not exist, with the exception of a few departments or research centres in bioinformatics of a certain stature. The opportunities for training and coordination among groups active in the field of Bioinformatics, located in the various research institutes, are almost non-existent and are entrusted to the will of the individual researchers.

In recent years various research themes in bioinformatics have been developed; they include:

• The mathematical modelling for a quantitative understanding of physical, chemical and biological phenomena agents.
• The hardware and software information technology for high performance and GRID computing.
• The development of new methods to tackle the problem of management and processing of large amounts of data.
• The development of data mining techniques for classification, multidimensional data analysis, identification and extraction of characteristic elements and of knowledge available from the data.

Regarding mathematical modelling, there are many biological problems at the molecular, cellular, and tissue level right up to complex organ simulation, which are formulated with mathematical modelling at different levels of abstraction and detail, using different methodologies, continuous and discrete mathematics, graph theory, systems and control theory, optimization systems and new formal languages to describe the complex systems, and even scientific visualization methods based on computer graphics techniques.

Regarding the classification of biological information, Machine Learning methodologies are widely used in this domain for their ability to extract information and build a model of a physical system from a set of experimental data on its behaviour.

This is a central problem in Bioinformatics sector: through current investigation technologies large amounts of data on a particular phenomenon have been generated, and it's extremely difficult to find an effective mathematical/statistical model that explains the observations made.

The same methods can provide essential support for the most advanced experimental techniques, such as gene expression analysis and Mass Spectrometry applied to proteomics, which generate significant amounts of data related to genes and proteins, otherwise difficult to analyse.

For this aim, the scientific literature of the sector represents a wide source of information for the validation and interpretation of the experiments. Hence the development of techniques able to analyse, classify and effectively find documents is a fundamental instrument of support to bioinformatics research.

As far as computational aspects are concerned, bioinformatics applications are "computationally-intensive", therefore their parallelization and implementation in contexts of heterogeneous resources based on Grid and Peer to Peer Computing is necessary.