Geometric deep learning reveals the spatiotemporal features of microscopic motion

Pineda, Jesús; Midtvedt, Benjamin; Bachimanchi, Harshith; Noé, Sergio; Midtvedt, Daniel; Volpe, Giovanni; Manzo, Carlo

doi:10.1038/s42256-022-00595-0

Cited by 31 publications

(20 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Noteworthy, the classification of the nodes of the network with respect to the corresponding scenario of formation would require development of new sprout-tracking algorithms. The fast development of machine-learning (ML) based tools 83, 84 promises possible applications of ML also in this field.…”

Section: Discussionmentioning

confidence: 99%

Topological evolution of sprouting vascular networks: from day-by-day analysis to general growth rules

Rojek,

Wrzos,

Żukowski

et al. 2023

Preprint

View full text Add to dashboard Cite

Engineering tissues with an embedded vasculature of well-controlled topology remains one of the basic problems in biofabrication. Still, little is known about the evolution of topological characteristics of vascular networks over time. Here, we perform a high-throughput day-by-day analysis of tens of microvasculatures that sprout from endothelial-cell coated micrometric beads embedded in an external fibrin gel. We use the bead-assays to systematically analyze (i) "macroscopic" observables such as the overall length and area of the sprouts, (ii) "microscopic" observables such as the lengths of segments or the branching angles and their distributions, as well as (iii) general measures of network complexity such as the average number of bifurcations per branch. We develop a custom angiogenic image analysis toolkit and track the evolution of the networks for at least 14 days of culture under various conditions, e.g., in the presence of fibroblasts or with added endothelial growth factor (VEGF). We find that the evolution always consists of three stages: (i) an inactive stage in which cells remain bound to the beads, (ii) a sprouting stage in which the sprouts rapidly elongate and bifurcate, and (iii) the maturation stage in which the growth slows down. We show that higher concentrations of VEGF lead to an earlier onset of sprouting and to a higher number of primary branches, yet without significantly affecting the speed of growth of the individual sprouts. We find that the mean branching angle is weakly dependent on VEGF and typically in the range of 60-75 degrees suggesting that, by comparison with the available Laplacian growth models, the sprouts tend to follow local VEGF gradients. Finally, we observe an exponential distribution of segment lengths, which we interpret as a signature of stochastic branching at a constant bifurcation rate (per unit branch length). Our results, due to high statistical relevance, may serve as a benchmark for predictive models and reveal how the external means of control, such as VEGF concentration, could be used to control the morphology of the vascular networks. We provide guidelines for the fabrication of optimized microvasculatures with potential applications in drug testing or regenerative medicine.

show abstract

Section: Discussionmentioning

confidence: 99%

Topological evolution of sprouting vascular networks: from day-by-day analysis to general growth rules

Rojek,

Wrzos,

Żukowski

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…This sharp spatial variation, introduced by binning, masks the precise underlying gradient of the diffusion coefficient change within a bin that may already be encoded in the data. Recent approaches, based on deep learning [14], remove the need for binning but require supervised training on labeled datasets.…”

Section: Introductionmentioning

confidence: 99%

“…To understand the dynamics of embedded membrane proteins, several methods have been proposed to map diffusion coefficients, or diffusivities, of membrane proteins [8, 9, 10, 11, 12, 13, 14].…”

Section: Introductionmentioning

confidence: 99%

Learning Continuous 2D Diffusion Maps from Particle Trajectories without Data Binning

Kumar,

Bryan,

Rojewski

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Diffusion coefficients often vary across regions, such as cellular membranes, and quantifying their variation can provide valuable insight into local membrane properties such as composition and stiffness. Toward quantifying diffusion coefficient spatial maps and uncertainties from particle tracks, we use a Bayesian method and place Gaussian Process (GP) Priors on the maps. For the sake of computational efficiency, we leverage inducing point methods on GPs arising from the mathematical structure of the data giving rise to non-conjugate likelihood-prior pairs. We analyze both synthetic data, where ground truth is known, as well as data drawn from live-cell single-molecule imaging of membrane proteins. The resulting tool provides an unsupervised method to rigorously map diffusion coefficients continuously across membranes without data binning.

show abstract

“…Higher control will enable more fundamental scientific discoveries about far-from-equilibrium phenomena, while being useful for applications, e.g., in sensing, nanomedicine, and materials science . Nowadays, the prospects for light actuation are ever brighter thanks to the development of several new technologies, such as cheaper lasers at all wavelengths, more versatile spatial light modulators, higher-speed cameras, and advanced particle tracking algorithms based on machine learning. − …”

Section: Introductionmentioning

confidence: 99%

Light, Matter, Action: Shining Light on Active Matter

Rey

Volpe

2023

ACS Photonics

Self Cite

View full text Add to dashboard Cite

Light carries energy and momentum. It can therefore alter the motion of objects on the atomic to astronomical scales. Being widely available, readily controllable, and broadly biocompatible, light is also an ideal tool to propel microscopic particles, drive them out of thermodynamic equilibrium, and make them active. Thus, light-driven particles have become a recent focus of research in the field of soft active matter. In this Perspective, we discuss recent advances in the control of soft active matter with light, which has mainly been achieved using light intensity. We also highlight some first attempts to utilize light's additional properties, such as its wavelength, polarization, and momentum. We then argue that fully exploiting light with all of its properties will play a critical role in increasing the level of control over the actuation of active matter as well as the flow of light itself through it. This enabling step will advance the design of soft active matter systems, their functionalities, and their transfer toward technological applications.

show abstract

Geometric deep learning reveals the spatiotemporal features of microscopic motion

Cited by 31 publications

References 45 publications

Topological evolution of sprouting vascular networks: from day-by-day analysis to general growth rules

Topological evolution of sprouting vascular networks: from day-by-day analysis to general growth rules

Learning Continuous 2D Diffusion Maps from Particle Trajectories without Data Binning

Light, Matter, Action: Shining Light on Active Matter

Contact Info

Product

Resources

About