Leggi o scarica la rivista completa
Copertina e Sommario
(2,16 MB PDF)
Sharing Ideas to Move Up
Luca Paolo Bolognini
Dr. Lisa Bartoli
BIOINFORMATICS: a Machine-Learning Approach to the prediction of protein structure, function and interactions
With the great advances in molecular biology and genomics we are now living what is called the “post-genomic era”, characterized by a massive increase in the amount of biological data available.
Bioinformatics is the application of computer technology and modeling expertise to the analysis and the management of biological data. The aim of Bioinformatics is to find out the meaning of biological information hidden in this huge amount of data and to deeply uncover the fundamental biology mechanisms that regulate all living cells. To understand the whole behavior of an organism is a very difficult issue, to be addressed at different levels of complexity including the study of proteomes and genomes.
In this paper the Bioinformatic approach to the study of protein structure and function is described. It consists in the design and implementation of computational models and tools that can provide insight in the function of proteins based on their sequence, their structure, their evolutionary history and their association with other molecules.
Machine-learning methods, such as Hidden Markov Models and Neural Networks, applied to the prediction of protein features will be presented. This interdisciplinary research area could have significant applications in human health, environment, food and energy fields.