Study of the use of the internal space of cells to expand the variety and efficiency of their computational capacity.

JUAN IGNACIO MARRONE; HERNAN GRECCO; ALEJANDRA C VENTURA

Buenos Aires

Conferencia; 27th IUPAP International Conference on Statistical Physics, StatPhys27; 2019

IUPAP C3 Commission, Statistical Physics

Throughout evolution, cells have developedcomplex systems, mainly protein systems, which allow them to utilize andprocess information. Our work has the goal of characterizing how it impacts inthe computational capacity of the cell, the spatial heterogeneity of itssignaling transmission systems. This idea brings together two concepts. First:the spatial heterogeneity, which causes the molecular partition of thosecomplex systems that process information. Second: the computational capacity ofthe cell, by which we mean, its ability to ?translate? environmental statesinto distinguishable internal states. Signaling networks spatially organizedare generally studied in relation to spatial process (e.g. polarization, chemotaxis,division and development), but surprisingly, in other cases, it seems the spaceis utilized as an additional computational dimension. In our work, we studied two computational models.The first one comes from (Roob III, Trendel, tenWolde and Mugler 2016), where they simulate how cooperative clusteringdigitizes biochemical signaling and enhances its fidelity, in a system ofactive Ras molecules at the cellular membrane. The higher the cooperativity,the higher the digitizing effect, and the lower the noise. The second modelcomes from (Fletcher, Clément, Vidal, Tabak and Bertram 2014), in which theystudy the steady-state response for ligand-bound receptors (RL) versus theamount of ligand (L). Then, they introduce the possibility of dimerization, andanalyze its response vs. L.With these twocomputational models, we proceeded to study what happens when RL can formhigher-sized clusters than a dimer, by modeling the reactions according to(Roob III et al. 2016). This means, introducing cooperativity in thedissociation reactions for clusters with a size higher than two, and taking asthe output, the amount of maximally sized clusters. This was done in both a deterministic(Fletcher et al. 2014) and a stochastic way (Roob III et al. 2016). Our goal isto find whether the digitization effect found in (Roob III et al. 2016) isinherent to the stochastic nature of the model, or if there is an analogouseffect when modeled in a deterministic way.