Italian Innovation

“The approach taken by DREAM”, adds Andrea Pinna, “is to organize a yearly competition that teams from all over the world can enter to present their algorithms”. "The teams taking part in the challenge receive datasets from experiments and need to predict results that are known only by the challenge organizers", explains Nicola Soranzo: "by checking how well the predictions achieved by the participants coincide with the actual results, we can better understand which algorithms are most useful to obtain reliable models from genomic datasets".

29 teams from all over the world, from top universities such as University of Chicago, Vanderbilt University, New York University and ParisTech, entered the DREAM5 network inference challenge. The challenge organizers gave the teams four datasets: three were real biological experiments (from bacteria and yeast) and one consisted of computer-simulated experiments. Each dataset consisted of measurements obtained from several experimental situations: i.e., underexpressing or overexpressing a gene, addition of drugs or introducing changes during a time series. The aim of the research was to identify the links between transcription factors and their target genes. "The challenge organizers", explains Vincenzo De Leo, the researcher from Linkalab who joined ALF team, "gave the teams two months to solve this task: we used this time to define the techniques to obtain the best results to answer the challenge questions".

Alberto de la Fuente, ALF team's scientific head, participates in DREAM competition from its second edition and already had been a best performer in DREAM challenges in 2007 and 2009. De la Fuente develops new methodologies for integrative analysis of gene expression and genotyping data (“systems genetics”) to elucidate gene networks. He also focuses on the identification of dysfunctional gene networks from disease gene expression data (“differential networking”). Very often complex diseases are due to the altered interaction between sets of genes, rather than from changes occurred to a single gene. Understanding how these dynamics work using network inference can mean a progress in the research against these diseases. With the words of de la Fuente, network inference from large-scale genomics data works like that: "A cause-effect relation between the genes is established, for example applying specific gene perturbations and measuring which genes change their expression level; these genes are causally affected by the perturbed genes. Then you have to distinguish between direct and indirect cause-effect relations. The resulting network model, a gene network, only consists of connections with direct cause-effect relations between the genes”.

Vincenzo De Leo works on a project in co-operation with CRS4 and funded by the Sardinian Region, applying techniques of differential networking analysis to data about diabetes in Sardinia. “Reconstructing complex gene networks is a very important step towards the comprehension of cellular behavior”, explains De Leo: “the amounts of data biologists gathered in the last few years typically consist of lists of biological components and information about their interactions: therefore, they are very suitable to be analyzed by Complex networks Theory”.

Nicola Soranzo is developing and evaluating computational methods to detect disease related changes in regulatory RNA networks.

Andrea Pinna is developing a software tool, SysGenSIM, to simulate gene expression and genotyping data with large gene networks. This software has been used to create the dataset for another challenge at the DREAM5, the Systems genetics challenge, and was crucially important to the good performance of the ALF team in the network inference challenge.

“Our simulator enabled us to test many different methods on data simulated according to the experiments of the DREAM5 Network Inference challenge,” explains Pinna, “so that we were able to select the most effective ones to apply to the challenge data”.

“Interestingly”, remarks Soranzo, “we trained our methods on our own simulated data and while we had a good performance on the real data, we didn’t have such a good performance on the simulated data in the challenge: this implies that our simulator generates data which better reflect real data than those simulated by the challenge organizers”.

For next year’s DREAM6 competition, the members of the ALF team will contribute again to a Systems Genetics challenge that will be based on data produced with SysGenSIM, and are planning to participate in one or more other DREAM6 challenges.

CRS4 Bioinformatics Laboratory

CRS4 Bioinformatics is the computational biology laboratory of the Center for Advanced Studies, Research and Development in Sardinia. The research is directed towards the development of new computational models for the identification of disease-associated markers using functional genomic and systems biology approaches.
Relevant examples of the current research include: data management and integration of high-throughput datasets for the development of models for personalized-medicine; identification of biomarker candidates associated with sample stratification, disease susceptibility or clinical outcome and systems biology: inference of gene and protein networks in complex biological systems of biomedical interest and design and implementation of mathematical algorithm for the modeling of biological processes.

Laboratory for the study of complex systems Linkalab

During the last 10 years, Complex Network Theory has proved to be a valid approach to the analysis of the most diverse phenomena: scientific, social and technological. Linkalab mainly deals with the application aspects of Complex Network Theory, with a special attention on the computational side, the optimization of calculation systems and the scientific cloud computing. With international quality standards, Linkalab developed theoretical and calculation tools in order to work as a research service for P.A.s and companies showing Complex Systems Analysis problems. Linkalab also develops web-groupware systems and last generation e-learning systems for knowledge management activities on web 2.0 platforms.

Further reading:

The website of the conference   

The website of the DREAM project   

The website of CRS4 Bioinformatica  

The website of Linkalab   

Dr. Alberto de la Fuente’s profile   

The poster describing the ALF team’s approach   

The abstract of the work   

The official certificate   

The ‘bow-tie’ layout of a yeast gene network obtained by the group at CRS4 Bioinformatica by integrative analysis of gene expression and genotype data. Layout created by Gianmaria Mancosu using the software Pajek.