Research themes



- Study of the dynamic and structural properties of DNA promoter sequences. (Livi, Pettinato, Di Patti)

We point out how the peculiar structural properties of promotor sequences (characterized by both ordered and disordered regions) influence the formation of breathers and solitons, focusing on energy transmission and localization processes. The statistical analysis of subsequences permitted to characterize transposons.

- Collective behaviors in networks of pulse coupled neural networks (Torcini - Olmi)

We have investigated analytically the linear stability of synchronous and asynchronous collective solutions in pulse coupled neural networks. We have determined in full generality the Floquet spectra for such solutions as a function of the pulse width, single neuron model characteristics and network size.

- Emergence of slow collective oscillations in neural networks with spike timing dependent plasticity (Torcini)

We have shown that the presence of spike timing dependent plasticity in pulse coupled neural networks can induce the emergence of slow collective oscillations in the frequency range of infra-slow oscillations. A novel mechanism, Sisyphus effect, has been identified capable to induce collective irregular oscillations in a deterministic model. this mechanism is induced by the feedback action of the level of synchronization among the neurons on the synaptic weights.

- Pattern formation (Fanelli, Di Patti, Cianci)

We have developed mathematical and numerical techniques to study the spontaneous emergence of patterns (Turing like or travelling waves) for reaction-diffusion models defined on symmetric networks. The emergence of coherent structures is revealed via a generalized Fourier transform, the signal being expanded on a basis formed by the eigenvectors of the Laplacian operators. A linear stability analysis can be performed which enables one to single out the region of parameters that yields the self-organized patterns. We have then developed the theory of patterns formation on directed networks (asymmetric graphs), as e.g. the map of the neural connection in the brain. Homogeneous fixed points can become unstable due to the topology of the network, resulting in a new class of instabilities which cannot be induced on undirected graphs. Beyond the deterministic reaction diffusion models, we have also analyzed their stochastic analogues. Each node of the network is imagined to host a large though finite number of individuals, e.g. neurons. Single individual effects, stemming from the endogenous discreteness of the analyzed medium, can prove crucial by modifying significantly the mean-field predictions, and can in particular induce the emergence of regular macroscopic patterns, both in time and space. The power spectrum of fluctuations can be analytically calculated by developing and systematizing the Linear Noise Approximation technique to network-based applications. The theory of stochastic patterns formation has been also developed for reaction diffusion systems defined on a regular lattices. We have then developed a general framework to address the stochastic theory of discrete time dynamical systems (maps). Another line of investigations has to to with the study of diffusive transport in densely populated spatial environment (molecular crowding).

- Cognitive Dynamics (Bagnoli)

We have developed cognitive-inspired methods for detecting communities in networks. We have also performed experiments with small groups of users interacting via chatlines, detecting and analyzing the communication networks for several conditions. We have developed a tri-partite model of cognitive activity and testing it on existing literature and via custom experiments.


A.Pluchino - A.Rapisarda - V.Latora

Current Research Topics:    

- Beneficial role of random strategies in financial markets (
- Noise and ether-drift experiments
- SOC models in Complex Networks with different topologies  
- Altruism in collective games on different networks 
- Agent based models simulations 
- Topological structure of multiplex networks
- Phase transition in growing systems
- Remote synchronization and network symmetries 
- City ecosystem resilience analysis
- Evolutionary dynamics of social interactions 

Home Links:


Diffusion and sensing of signal molecules in bacterial colonies.

Quorum Sensing is a mechanism widely used by bacterial cells to modify their
gene expression pattern, underlying many known instances of collective
behaviour in bacterial colonies. The mechanism is based on sensing the
concentration of small diffusible signal molecules that are produced by
bacteria themselves. Single cells can thus "measure"
how many neighbours they have and collectively switch to a different behaviour
if some "quorum" threshold is reached.

Complex feedback mechanisms can emerge, as different bacterial cells can
produce and sense different signal molecules. Notably, quorum sensing
can be engineered to "program" bacterial colonies as operating light-sensing
computers where different algorithms can be implemented [Tabor et al, Cell 137:
1272-1281, 2009].

We plan to study the diffusion of signal molecules in bacterial
colonies as a standard diffusion problem. In particular, we will address
the role played by other factors than bacterial density, such as the nature of
boundaries and the size and geometry of the colony, in governing colony
behavior. The theoretical modeling will be developed in close
connected with experiments run by Prof. A. Squartini at DAFNAE (Unipd).

1) Lo scambio di cibo tra nazioni puo' essere tradotto in una stessa unita' di misura
in termini di costo in acqua necessaria , acqua virtuale, per produrlo.  Studieremo
le reti di flussi di acqua virtuale tra nazioni e come questa influenzi la crescita
demografica delle nazioni stesse utilizzando equazioni stocastiche [1].

2) Le reti di interazione tra le specie in un sistema ecologico mostrano
architetture cosiddette "nested". Queste si osservano anche in reti di interazione
sociali/economiche. Cercheremo di capire se queste strutture ricorrenti derivino da
qualche principio di ottimizzazione/variazionale legato all'interazione di mutuo
vantaggio che si verifica nelle reti suddette [2].

1. S. Suweis, A. Rinaldo, A. Maritan , P. D'Odorico, Proc. Natl. Acad. Sci.
(USA), 10, 92-95 (2013).

2. Bascompte, J. & Jordano, P. Plant-Animal Mutualistic Networks: The Architecture
of Biodiversity. Annual Review of Ecology, Evolution, and Systematics 38,
567–593 (2007).

Earthquakes are a typical phenomenon extremely far from equilibrium,
displaying complex spatio-temporal patterns that are phenomenologically
well characterized but not understood yet. There are models for
earthquakes where each event can be seen as a unit that generates
off-springs according to its magnitude and to known correlations between
events. Of course these models cannot reproduce the full complexity of
seismicity, but the synthetic catalogs they generate are often
considered a good benchmark for testing algorithms for aftershock or
foreshock detenction. There is however still a lot of room for improving
these models and our general understanding of earthquakes.

Methods of the physics of complex systems applied to quantitative finance.
Modeling of asset dynamics inspired by the renormalization group approach
to critical phenomena. Option pricing beyond Black-Scholes based on
closed formulas.
vedi: arXiv:1305.3243

Anomalous dynamics of polymers, especially translocation.

Active matter models (Enzo Orlandini).