Discovering the secrets of cells: A bioinformatician’s dream

DREAM is an international scientific project with the aim to improve methods for studying molecular interactions from a network oriented perspective. An important goal in modern biology is to understand molecular-level processes that occur inside cells: how does a gene’s activity affect the activity of other genes? What kind of influence does a gene express on a protein’s activity? How does a protein regulate the speed of cellular reactions? In addition, researchers analyze the behavior of cellular “players” during changes in external conditions, thereby elucidating their precise roles.

The very small space of the cell's interior has a high density of thousands of different types of molecules (for example, DNA, RNA, proteins and sugars) that continuously participate in chemical reactions. Modern experimental techniques enable us to measure a large number of these molecules simultaneously. Since dealing with such large datasets is very complicated, the use of mathematical methods are certainly needed. For example, in order to elucidate interactions between genes or proteins one can use the method of fitting to the data a mathematical model that describes a network of interactions, an approach often called ‘reverse-engineering’.

In order to stimulate discussion between the various research groups that are developing mathematical techniques and software for analyzing bio-molecular interaction networks, the DREAM project organizers setup a competition. The competition would reveal the strengths and weaknesses of the various methods, at least with respect to the specific kinds of datasets used in the competition. The competition was divided into five major categories, in turn divided into 34 subcategories. In total 35 groups from all over the world participated. The competitors had to reconstruct interaction networks between genes or proteins, using the data given by challenge organizers. The goal was to produce the result closest to the real network structures, known only to the competition organizers.

The RAGNO group, from CRS4 Bioinformatica, took part in the competition. Alberto de la Fuente, the team coordinator, said that competing with groups from different countries and different scientific fields has been exciting: RAGNO ranked first place in six categories. Alberto de la Fuente said that: “reverse engineering complex gene networks is a crucial step in understanding how our cells function. By taking part in the competition we also demonstrated that the life of a researcher is not boring as many people think, but it can be fun and such challenges can promote collaboration between scientists, necessary to reach a common goal”.

In the competition the RAGNO team used new and innovative analysis tools, based on original methods implemented by, besides de la Fuente, Giorgio Fotia, Fabio Maggio and Wieslawa Mentzen from the Bioinformatics Laboratory, Alan Scheinine from CRS4 and Angela Baralla, student of master in "Computational and systems biology" of the Microsoft Research Center and University of Trento and trainee at CRS4 Bioinformatica under the supervision of Alberto de la Fuente. Only one other Italian group participated in the competition: Diego di Bernardo’s group at TIGEM, who also had a winning submission.