Collective decision-making in social animal groups often emerges from simple local interactions and can give rise to abrupt transitions between distinct behavioral states. In this talk, we explore how coordination, information transfer, and environmental constraints shape these transitions in animal collectives. Using controlled experimental systems combined with quantitative analysis, we investigate how groups integrate sensory information and social interactions to reach consensus and coordinate movement, particularly in constrained or heterogeneous environments.

Our results highlight the key role of group interactions and sensorial cues in driving collective outcomes, revealing how small changes in environmental conditions or information accessibility can trigger qualitative shifts in group behavior, akin to phase transitions in physical systems. We further discuss how the interplay between individual behavior, group cohesion, and environmental structure governs the efficiency of collective decisions and navigation. These findings contribute to a broader understanding of active and living matter, bridging physics and biology, and provide new perspectives on how collective intelligence emerges in natural systems.

From subcellular molecular networks to ecological communities, how living systems respond to external stimuli is a core aspect of biological function.  Here, theoretical modeling built on physical principles can help us understand the underlying mechanisms.  In this talk, I discuss two such examples.  The first is a nonequilibrium limit to how sensitively a biochemical system can respond to an external perturbation.  To illustrate this result, I will draw on examples from biophysics, where the effectiveness of numerous biochemical systems depends on being exquisitely sensitive to changes in chemical inputs.  We will see how these predictions rationalize known energetic requirements of some common biochemical motifs and provide new limits to others.  The second is a study of how cooperativity and spatial structure in bacterial biofilms conspire to lead to a community-level antibiotic resistance.  I will present experiments measuring the population dynamics of coupled spatially fixed biofilms and planktonic populations of a mixture of drug-sensitive and resistant E. faecalis.  Matched with theoretical modeling, we surprisingly find no spatial structure in the biofilm even though there is a common population inversion, in which the final fraction of resistant cells exceeds its initial value, particularly at smaller initial resistant fractions.

Protein generative models are often viewed as tools for designing novel sequences, but they can also be interpreted more fundamentally as data-driven maps capturing evolutionary constraints. In this perspective, sequence probabilities define an effective evolutionary landscape that links mutational effects, epistasis, and long-term protein diversification within a common framework. This talk discusses how models learned from natural sequence variation can move beyond generation and prediction to provide an interpretable description of protein evolution across scales, from single mutations in their local sequence context to the emergence of distant homologs.

The aging process represents one of biology's most complex system-level phenomena. A major challenge is moving from observing its correlates to identifying its fundamental, targetable bottlenecks. In this talk, I will explore a reverse-engineering approach, using pharmacological interventions in model organisms to deconstruct the mechanisms of aging and pinpoint promising avenues for intervention.  I will discuss how we can leverage existing biological data and what new, targeted measurements are required to fill critical gaps. A key question is the selection of appropriate model organisms that offer the right balance of physiological complexity, experimental tractability, and translational relevance for aging research. Furthermore, I will examine the role of artificial intelligence in this endeavor: while AI excels at finding generalizable patterns, its success is critically dependent on the quality and nature of the underlying data—an area where significant improvements are needed.  I will present examples from my work across multiple species, including the development of a scalable high-throughput platform for pharmaco-biology in Daphnia. This system allows us to characterize drug-induced perturbations and link them to lifespan and healthspan outcomes. We will discuss a computational framework to regress macro-phenotypes to the molecular pathways. Finally, I will outline central challenges in the field and propose concrete directions for researchers interested in joining the effort to reverse engineer aging.

Plant development occurs as an interplay of signalling, growth, and environmental cues, becoming a highly dynamic and complex system. In our lab, we study the dynamics of plant developing tissues, combining mathematical modelling, confocal microscopy, and quantitative image analysis.In this talk, I will first explain how cells become different from one another as they divide and grow in the Arabidopsis plant sepal, focusing on the case of giant cell patterning. I will demonstrate that giant cells appear in a random spatial pattern and become clustered while the surrounding cells divide. This finding shows that cell proliferation can have a fundamental role in shaping spatial patterns. Secondly, I will present our work on the dynamics of the floral transition in Arabidopsis, by which the plant changes from producing leaves to producing flowers. Our results indicate that a time-dependent bistable switch underlies the floral transition dynamics, leading to a critical slowing down and introducing an unexpected timescale on the transition. I will also discuss how this behaviour relates to the transition robustness and its reversibility.