Statistical Inference and Computational Biology

Definition: Development of efficient statistical-physics inspired techniques for inference and optimization, and application to the analysis of large-scale experimental data and to the inference in complex biological systems.

Innovative aspect: Analyzing optimally the large amount of experimental data accumulated by modern sequencing and other high-throughput techniques frequently leads to computational problems, which are intractable by conventional algorithmic approaches. However, the last years have seen the development of novel and highly efficient optimization and inference techniques inspired by the statistical physics of complex disordered systems. First applications to biological inference emerge as a proof-of-concept of the possibility to transfer these techniques to computational biology.

Added value: These new computational techniques shall be brought to the full benefit of biological research, for the quantitative analysis of large-scale biological data (gene-expression, DNA-copy number, evolutionary sequence variability of homologous proteins) and the inference of complex biological networks for protein-protein interactions, signal transduction and genetic interactions.

Ongoing Projects:
1. Algorithms of large-scale inference problems in molecular biology and phylogeny
2. Inference of protein-protein interactions from multi-species sequence data
3. Inference of signaling networks from multiple-perturbation experiments on cancer cell lines
4. Association studies and integrated data analysis

Head of Unit: Andrea Pagnani
Phone +39 0116706488

Riccardo Zecchina, Senior Researcher -
Marco Zamparo, Researcher -

Alfredo Braunstein -
Alessandro Pelizzola
Martin Weigt -

Carlo Baldassi -
Abolfazl Ramezanpour -
Carla Bosia - Quantitative Biology Unit co-PI -

Indaco Biazzo -

2013 Publications

2012 Publications

2011 Publications