arXiv subjects

Giorgio Gosti

Publications and source records attributed to Giorgio Gosti.

18 recordsLinked to original sources

arXiv2026

Inheritance Entropy: A Model-Independent Method to Probe the Hereditary Structure of Cell Lineage Trees

Human bone marrow stromal cells (BMSC) include skeletal stem cells with ground-breaking therapeutic potential. However, BMSC colonies have very heterogeneous in vivo behaviour, due to their different potency; this unpredictability is the greatest hurdle to the development of skeletal regeneration therapies. Colony-level heterogeneity urges a fundamental question: how is it possible that one colony as a collective unit behaves differently from another one? If cell-to-cell variability were just an uncorrelated random process, a million cells in a transplant-bound colony would be enough to yield statistical homogeneity, hence washing out any colony-level traits. A possible answer is that the differences between two originating cells are transmitted to their progenies and collectively persist through an hereditary mechanism. But non-genetic inheritance remains an elusive notion, both at the experimental and at the theoretical level. Here, we prove that heterogeneity in the lineage topology of BMSC clonal colonies is determined by heritable traits that regulate cell-cycle exit. The cornerstone of this result is the definition of a novel entropy of the colony, which measures the hereditary ramifications in the distribution of inactive cells across different branches of the proliferation tree. We measure the entropy in 32 clonal colonies, obtained from single-cell lineage tracing experiments, and show that in the greatest majority of clones this entropy is decisively smaller than that of the corresponding non-hereditary lineage. This result indicates that hereditary epigenetic factors play a major role in determining cycle exit of bone marrow stromal cells.

physics.bio-ph↗

arXiv2026

VertAX: a differentiable vertex model for learning epithelial tissue mechanics

Epithelial tissues dynamically reshape through local mechanical interactions among cells, a process well captured by vertex models. Yet their many tunable parameters make inference and optimization challenging, motivating computational frameworks that flexibly model and learn tissue mechanics. We introduce VertAX, a differentiable JAX-based framework for vertex-modeling of confluent epithelia. VertAX provides automatic differentiation, GPU acceleration, and end-to-end bilevel optimization for forward simulation, parameter inference, and inverse mechanical design. Users can define arbitrary energy and cost functions in pure Python, enabling seamless integration with machine-learning pipelines. We demonstrate VertAX on three representative tasks: (i) forward modeling of tissue morphogenesis, (ii) mechanical parameter inference, and (iii) inverse design of tissue-scale behaviors. We benchmark three differentiation strategies-automatic differentiation, implicit differentiation, and equilibrium propagation-showing that the latter can approximate gradients using repeated forward, adjoint-free simulations alone, offering a simple route for extending inverse biophysical problems to non-differentiable simulators with limited additional engineering effort.

cs.LG↗

arXiv2025

Emergent learning: neuromorphic photonic computing with accelerated training

Emergent learning transforms a disordered optical medium into a photonic device capable of storage, recognition, and classification of arbitrary memory patterns. First, we show that the intensity at the output of a multiply scattering system can be described by a dyadic matrix, the optical-synaptic matrix, exhibiting the same form as a Hebbian synaptic matrix containing a single memory. Then, we employ emergent learning - an approach inspired by neuroscience - to exploit the vast dictionary of raw memories inherently available within a disordered optical structure, thereby engineering the optical-synaptic matrix to store a user-defined attractor, or tailored memory. Importantly these photonic structures also works as an optical comparators providing an intensity-based measure of the degree of similitude between a query pattern and the stored pattern, realizing an hardware co-localization between memory and optical operator. Our system has an almost infinite hardware capacity of tailored memories/ operators ($\mathcal{M} \sim 10^{60557}$), thus these tailored memories can be then employed as examples to build a classifier hardware based on intensity comparison without the need of additional digital transformation layers. Remarkably, this Photonic Emergent Learning platform is not only flexible and fabrication-free, but also relies primarily on analog processes, thus shifting the computational burden of training from the digital layers to the optical domain reducing the computational cost and enhancing performance.

physics.optics↗

arXiv2025

Lineage topology, replication kinetics and cell cycle synchronization reveal regulated growth dynamics in human bone marrow stromal cell colonies

Bone marrow stromal cells (BMSC) -- which include skeletal stem cells -- are a promising tool in regenerative medicine. However, their heterogeneous and unpredictable in vivo behaviour remains a critical barrier preventing the development of standardized therapeutic approaches for skeletal tissue regeneration. Several studies have attempted to identify in vitro features that could correlate with the in vivo differentiation properties, yet the mechanisms ruling BMSC heterogeneity remain poorly understood. Here, using time-lapse imaging, we lineage-trace 32 single-cell-derived BMSC colonies through seven generations. We observe significant inter-colony and intra-colony heterogeneity in lineage topology (determined by the number of senescent or apoptotic cells) and in replicative kinetics (measured from proliferating cells only). Interestingly, topology and kinetics result strongly correlated, suggesting the existence of regulatory factors linking the non-dividing/apoptotic subpopulations with proliferating cells. Furthermore, BMSCs display highly synchronized cell cycles during early generations, indicating stage-specific regulatory mechanisms through which cells influence each other. By employing a non-interacting population growth model, we demonstrate that the observed synchronisation cannot be explained by an uncorrelated branching process; instead, cell-to-cell correlation of division times must exist. Our findings reveal fundamental mechanisms governing BMSC heterogeneity and growth dynamics that may inform strategies to control their regenerative potential.

q-bio.CB↗

arXiv2025

Robust assessment of asymmetric division in colon cancer cells

Asymmetric partition of fate determinants during cell division is a hallmark of cell differentiation. Recent work suggested that such a mechanism is hijacked by cancer cells to increase both their phenotypic heterogeneity and plasticity and in turn their fitness. To quantify fluctuations in the partitioning of cellular elements, imaging-based approaches are used, whose accuracy is limited by the difficulty of detecting cell divisions. Our work addresses this gap proposing a general method based on high-throughput flow cytometry measurements coupled with a theoretical framework. We applied our method to a panel of both normal and cancerous human colon cells, showing that different kinds of colon adenocarcinoma cells display very distinct extents of fluctuations in their cytoplasm partition, explained by an asymmetric division of their size. To test the accuracy of our population-level protocol, we directly measure the inherited fractions of cellular elements from extensive time-lapses of live-cell laser scanning microscopy, finding excellent agreement across the cell types. Ultimately, our flow cytometry-based method promises to be accurate and easily applicable to a wide range of biological systems where the quantification of partition fluctuations would help accounting for the observed phenotypic heterogeneity and plasticity

q-bio.CB↗

arXiv2025

A mother-machine microfluidic device for non-adherent mammalian cells reveals the population growth strategies

We develop a mother machine-like microfluidic device specifically designed to track the proliferation of T-cells via live-cell microscopy. Although numerous microfluidic setups have been developed to study cell proliferation at the single-cell level, most of them are optimized for use on adherent cells. Here, we present a device to track the proliferation of suspension cells, featuring an array of microchannels that trap cells, easing their monitoring while allowing for controlled growth conditions. Each microchannel, whose geometry has been optimized through computational fluid dynamics simulations, allows a single cell to enter and proliferate while maintaining a continuous flow of nutrients, ensuring long-term monitoring over multiple generations. We show the advantages of this system in characterizing the proliferation of human leukemia T-cells. In particular, we follow the growth and division over multiple generations, finding that cells exhibit a slightly asymmetric volume division where deviations in the size are compensated by a size-like division strategy. Overall, our device design can be easily adapted and used to study different cell types and sizes while maintaining the same high trapping efficiency.

q-bio.CB↗

arXiv2025

Rosette formations as symmetry-breaking events: theory and experiment

Multicellular rosettes are observed in different situations such as morphogenesis, wound healing, and cancer progression. While some molecular insights have been gained to explain the presence of these assemblies of five or more cells around a common center, what are the tunable, global features that favors/hinders their formation is still largely unknown. Here, we made use of a Voronoi dynamical model to investigate the ingredients driving the emergence of rosettes characterized by different degree of stability and organization. We found that (i) breaking the local spatial symmetry of the system, i.e., introducing curvature-inducing defects, allows for the formation of rosette-like structures (ii) whose probability of formation depends on the characteristic of the cellular layer. In particular, a trade-off between tissue fluidity and single cell deformability dictates the assembly of transient rosettes, that are strongly stabilized in the presence of cell alignment interactions. To test our model predictions, we performed fluorescence microscopy experiments on rosette-forming neural populations derived from induced pluripotent stem cells, finding significant agreement. Overall, our work may set the stage to gain an unifying understanding of the plethora of biophysical mechanisms involving the occurrence of rosette-like structure both in physiology and their altered formation in pathology.

q-bio.CB↗

arXiv2025

Evidence of Scaling Regimes in the Hopfield Dynamics of Whole Brain Model

It is shown that a Hopfield recurrent neural network exhibits a scaling regime, whose specific exponents depend on the number of parcels used and the decay length of the coupling strength. This scaling regime recovers the picture introduced by Deco et al., according to which the process of information transfer within the human brain shows spatially correlated patterns qualitatively similar to those displayed by turbulent flows, although with a more singular exponent, 1/2 instead of 2/3. Both models employ a coupling strength which decays exponentially with the Euclidean distance between the nodes, informed by experimentally derived brain topology. Nevertheless, their mathematical nature is very different, Hopf oscillators versus a Hopfield neural network, respectively. Hence, their convergence for the same data parameters, suggests an intriguing robustness of the scaling picture.Furthermore, the present analysis shows that the Hopfield model brain remains functional by removing links above about five decay lengths, corresponding to about one sixth of the size of the global brain. This suggests that, in terms of connectivity decay length, the Hopfield brain functions in a sort of intermediate ``turbulent liquid''-like state, whose essential connections are the intermediate ones between the connectivity decay length and the global brain size. The evident sensitivity of the scaling exponent to the value of the decay length, as well as to the number of brain parcels employed, leads us to take with great caution any quantitative assessment regarding the specific nature of the scaling regime.

cond-mat.dis-nn↗

arXiv2024

Probing leukemia cells behavior under starvation

The ability of a cancer cell population to achieve heterogeneity in their phenotype distributions offers advantages in tumor invasiveness and drug resistance. Studying the mechanisms behind such observed heterogeneity in mammalian cells presents challenges due for instance to the prolonged proliferation times compared to widely studied unicellular organisms like bacteria and yeast. Here, we studied the response of leukemia cell populations to serum starvation via a protocol, we recently developed, that makes use of live cell fluorescence and flow cytometry in combination with a quantitative analytical model to follow the population proliferation while monitoring the dynamics of its phenotype distributions. We found that upon switching between a serum-rich to a serum-poor media, leukemia cells (i) maintain a memory of the previous environment up to one generation even in the presence of severe medium-depletion, before (ii) adapting their growth and division rates to the novel environment while preserving a sizer-like division strategy. Finally, looking at the mitochondria content of the proliferating vs non-proliferating cells, we found that the latter is characterized by a higher number of older mitochondria, suggesting a possible functional role of the observed asymmetric partitioning of (aged) mitochondria in leukemia cells.

q-bio.CB↗

arXiv2023

Determining cancer cells division strategy

Heterogeneity in the size distribution of cancer cell populations has been recently linked to drug resistance and invasiveness. However, despite many progresses have been made in understanding how such heterogeneous size distributions arise in fast-proliferating cell types -like bacteria and yeast-, comprehensive investigations on cancer cell populations are still lacking mainly due to the difficulties of monitoring the proliferation of the time scales typical of mammalian cells. From a reductionist cell dynamics point of view, the strategies allowing size homeostasis are roughly grouped into three classes, \emph{i.e.} timer, sizer, or adder. These strategies are empirically distinguishable given the phenomenological measurable relationship between the cell size at birth and at division, which requires following the proliferation at the single-cell level. Here, we show how it is possible to infer the growth regime and division strategy of leukemia cell populations using live cell fluorescence labeling and flow cytometry in combination with a quantitative analytical model where both cell growth and division rates depend on powers of the cell size. Using our novel approach, we found that the dynamics of the size distribution of leukemia Jurkat T-cells is quantitatively reproduced by (i) a sizer-like division strategy, with (ii) division times following an Erlang distribution given by the sum of at least three independent exponentially-distributed times and (iii) fluctuations up to 15\% of the inherited fraction of size at division with respect to the mother cell size. Finally, we note that our experimental and theoretical apparatus can be easily extended to other cell types and environmental conditions, allowing for a comprehensive characterization of the growth and division model different cells can adopt.

q-bio.CB↗

arXiv2023

Photonic Stochastic Emergent Storage: Exploiting Scattering-intrinsic Patterns for Programmable Deep Classification

Disorder is a pervasive characteristic of natural systems, offering a wealth of non-repeating patterns. In this study, we present a novel storage method that harnesses naturally-occurring random structures to store an arbitrary pattern in a memory device. This method, the stochastic emergent storage (SES), builds upon the concept of emergent archetypes, where a training set of imperfect examples (prototypes) is employed to instantiate an archetype in an Hopfield-like network through emergent processes. We demostrate this non-Hebbian paradigm in the photonic domain by utilizing random transmission matrices, which govern light scattering in a white-paint turbid medium, as prototypes. Through the implementation of programmable hardware, we successfully realize and experimentally validate the capability to store an arbitrary archetype and perform classification at the speed of light. Leveraging the vast number of modes excited by mesoscopic diffusion, our approach enables the simultaneous storage of thousands of memories without requiring any additional fabrication efforts. Similar to a content addressable memory, all stored memories can be collectively assessed against a given pattern to identify the matching element. Furthermore, by organizing memories spatially into distinct classes, they become features within a higher-level categorical (deeper) optical classification layer.

cond-mat.dis-nn↗

arXiv2022

Electrostatic complementarity at the interface drives transient protein-protein interactions

Understanding the molecular mechanisms driving the binding between bio-molecules is a crucial challenge in molecular biology. In this respect, characteristics like the preferentially hydrophobic composition of the binding interfaces, the role of van der Waals interactions (short range forces), and the consequent shape complementarity between the interacting molecular surfaces are well established. However, no consensus has yet been reached on how and how much electrostatic participates in the various stages of protein-protein interactions. Here, we perform extensive analyses on a large dataset of protein complexes for which both experimental binding affinity and pH data were available. We found that (i) although different classes of dimers do not present marked differences in the amino acid composition and charges disposition in the binding region, (ii) homodimers with identical binding region show higher electrostatic compatibility with respect to both homodimers with non-identical binding region and heterodimers. The level of electrostatic compatibility also varies with the pH of the complex, reaching the lowest values for low pH. Interestingly, (iii) shape and electrostatic complementarity behave oppositely when one stratifies the complexes by their binding affinity. Conversely, complexes with low values of binding affinity exploit Coulombic complementarity to acquire specificity, suggesting that electrostatic complementarity may play a greater role in transient (or less stable) complexes. In light of these results, (iv) we provide a fast and efficient method to measure electrostatic complementarity without the need of knowing the complex structure. Expanding the electrostatic potential on a basis of 2D orthogonal polynomials, we can discriminate between transient and permanent protein complexes with an AUC of the ROC of 0.8.

q-bio.BM↗

arXiv2021

Alignment interactions drive structural transitions in biological tissues

Experimental evidence shows that there is a feedback between cell shape and cell motion. How this feedback impacts the collective behavior of dense cell monolayers remains an open question. We investigate the effect of a feedback that tends to align the cell crawling direction with cell elongation in a biological tissue model. We find that the alignment interaction promotes nematic patterns in the fluid phase that eventually undergo a non-equilibrium phase transition into a quasi-hexagonal solid. Meanwhile, highly asymmetric cells do not undergo the liquid-to-solid transition for any value of the alignment coupling. In this regime, the dynamics of cell centers and shape fluctuation show features typical of glassy systems.

cond-mat.soft↗

arXiv2021

Prediction of gene expression time series and structural analysis of gene regulatory networks using recurrent neural networks

Methods for time series prediction and classification of gene regulatory networks (GRNs) from gene expression data have been treated separately so far. The recent emergence of attention-based recurrent neural networks (RNN) models boosted the interpretability of RNN parameters, making them appealing for the understanding of gene interactions. In this work, we generated synthetic time series gene expression data from a range of archetypal GRNs and we relied on a dual attention RNN to predict the gene temporal dynamics. We show that the prediction is extremely accurate for GRNs with different architectures. Next, we focused on the attention mechanism of the RNN and, using tools from graph theory, we found that its graph properties allow to hierarchically distinguish different architectures of the GRN. We show that the GRNs respond differently to the addition of noise in the prediction by the RNN and we relate the noise response to the analysis of the attention mechanism. In conclusion, this work provides a a way to understand and exploit the attention mechanism of RNN and it paves the way to RNN-based methods for time series prediction and inference of GRNs from gene expression data.

physics.bio-ph↗

arXiv2021

New algorithm to determine a minimal representation of the molecular surface

Most proteins perform their biological function by interacting with one or more molecular partners. In this respect, characterizing the features of the molecular surface, especially in the portions where the interaction takes place, turned out to be a crucial step in the investigation of the mechanisms of recognition and binding between molecules. Predictive methods often rely on extensive samplings of molecular patches with the aim to identify hot spots on the surface. In this framework, analysis of large proteins and/or many molecular dynamics frames is often unfeasible due to the high computational cost. Thus, finding optimal ways to reduce the number of points to be sampled maintaining the biological information carried by the molecular surface is pivotal. Here, we present a new theoretical and computational algorithm with the aim of determining a subset of surface points, appropriately selected in space, in order to maximize the information of the overall shape of the molecule by minimizing the number of total points. We test our procedure by looking at the local shape of the surface through a recently developed method based on the formalism of Zernike polynomials in two dimensions, which is able to characterize the local shape properties of portions of molecular surfaces. The results of this method show that a remarkably higher ability of this algorithm to reproduce the information of the complete molecular surface compared to uniform random sampling.

physics.bio-ph↗

arXiv2020

Does blood type affect the COVID-19 infection pattern?

Among the many aspects that characterize the COVID-19 pandemic, two seem particularly challenging to understand: (i) the great geographical differences in the degree of virus contagiousness and lethality which were found in the different phases of the epidemic progression, and (ii) the potential role of the infected people's blood type in both the virus infectivity and the progression of the disease. A recent hypothesis could shed some light on both aspects. Specifically, it has been proposed that in the subject-to-subject transfer SARS-CoV-2 conserves on its capsid the erythrocytes' antigens of the source subject. Thus these conserved antigens can potentially cause an immune reaction in a receiving subject that has previously acquired specific antibodies for the source subject antigens. This hypothesis implies a blood type-dependent infection rate. The strong geographical dependence of the blood type distribution could be, therefore, one of the factors at the origin of the observed heterogeneity in the epidemics spread. Here, we present an epidemiological deterministic model where the infection rules based on blood types are taken into account and compare our model outcomes with the exiting worldwide infection progression data. We found an overall good agreement, which strengthens the hypothesis that blood types do play a role in the COVID-19 infection.

q-bio.PE↗

arXiv2020

In-Silico evidence for two receptors based strategy of SARS-CoV-2

We propose a novel numerical method able to determine efficiently and effectively the relationship of complementarity between portions of proteins surfaces. This innovative and general procedure, based on the representation of the molecular iso-electron density surface in terms of 2D Zernike polynomials, allows the rapid and quantitative assessment of the geometrical shape complementarity between interacting proteins, that was unfeasible with previous methods. We first tested the method with a large dataset of known protein complexes obtaining an overall area under the ROC curve of 0.76 in the blind recognition of binding sites and then applied it to investigate the features of the interaction between the Spike protein of SARS-Cov-2 and human cellular receptors. Our results indicate that SARS-CoV-2 uses a dual strategy: its spike protein could also interact with sialic acid receptors of the cells in the upper airways, in addition to the known interaction with Angiotensin-converting enzyme 2.

physics.bio-ph↗

arXiv2018

Effect of dilution in asymmetric recurrent neural networks

We study with numerical simulation the possible limit behaviors of synchronous discrete-time deterministic recurrent neural networks composed of N binary neurons as a function of a network's level of dilution and asymmetry. The network dilution measures the fraction of neuron couples that are connected, and the network asymmetry measures to what extent the underlying connectivity matrix is asymmetric. For each given neural network, we study the dynamical evolution of all the different initial conditions, thus characterizing the full dynamical landscape without imposing any learning rule. Because of the deterministic dynamics, each trajectory converges to an attractor, that can be either a fixed point or a limit cycle. These attractors form the set of all the possible limit behaviors of the neural network. For each network, we then determine the convergence times, the limit cycles' length, the number of attractors, and the sizes of the attractors' basin. We show that there are two network structures that maximize the number of possible limit behaviors. The first optimal network structure is fully-connected and symmetric. On the contrary, the second optimal network structure is highly sparse and asymmetric. The latter optimal is similar to what observed in different biological neuronal circuits. These observations lead us to hypothesize that independently from any given learning model, an efficient and effective biologic network that stores a number of limit behaviors close to its maximum capacity tends to develop a connectivity structure similar to one of the optimal networks we found.

cond-mat.dis-nn↗