Comparison of complex systems: 'similarity'/'distance' measures

Comparison of biological objects has long been a rich source of new information. They are used in diverse domains to reconstruct evolutionary histories, to estimate the pertinence of animal models, or to compare healthy and diseased individuals.

Comparison of protein sequences and 3D structures is widely used to identify (i) conserved elements that often play important roles in the shared structure or function of a protein family and (ii) specific elements that may be responsible for the observed functional differences.

The methodologies that have been developed in this context have been applied to some extent to the comparison of protein interaction networks, but many challenges remain. Quantitative measures of the global similarity/distance between networks will be crucial, as well as efficient methods for the identification of locally conserved sub-systems.