Learning physical properties of anomalous random walks using graph neural networks

Abstract

Single particle tracking allows probing how biomolecules interact physically with their natural environments. A fundamental challenge when analysing recorded single particle trajectories is the inverse problem of inferring the physical model or class of models of the underlying random walks. Reliable inference is made difficult by the inherent stochastic nature of single particle motion, by experimental noise, and by the short duration of most experimental trajectories. Model identification is further complicated by the fact that main physical properties of random walk models are only defined asymptotically, and are thus degenerate for short trajectories. Here, we introduce a new, fast approach to inferring random walk properties based on graph neural networks (GNNs). Our approach consists in associating a vector of features with each observed position, and a sparse graph structure with each observed trajectory. By performing simulation-based supervised learning on this construct [1], we show that we can reliably learn models of random walks and their anomalous exponents. The method can naturally be applied to trajectories of any length. We show its efficiency in analysing various anomalous random walks of biological relevance that were proposed in the AnDi challenge [2]. We explore how information is encoded in the GNN, and we show that it learns relevant physical features of the random walks. We furthermore evaluate its ability to generalize to types of trajectories not seen during training, and we show that the GNN retains high accuracy even with few parameters. We finally discuss the possibility to leverage these networks to analyse experimental data.

References in corpus (44)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Semi-Supervised Classification with Graph Convolutional Networks
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• WaveNet: A Generative Model for Raw Audio
• Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
• Fast Graph Representation Learning with PyTorch Geometric
• Scale-free correlations in bird flocks
• The frontier of simulation-based inference
• Variational Graph Auto-Encoders
• Hierarchical Graph Representation Learning with Differentiable Pooling
• Lévy walks
• Ergodic Properties of Fractional Brownian-Langevin Motion
• Quantitative analysis of single particle trajectories: mean maximal excursion method
• Nonergodic Subdiffusion from Brownian Motion in an Inhomogeneous Medium
• Scaled Brownian motion: a paradoxical process with a time dependent diffusivity for the description of anomalous diffusion
• Probing microscopic origins of confined subdiffusion by first-passage observables
• Nonparametric Bayesian inference of the microcanonical stochastic block model
• Mean first-passage times of non-Markovian random walkers in confinement
• Classification of diffusion modes in single-particle tracking data: Feature-based versus deep-learning approach
• Differentiable Graph Module (DGM) for Graph Convolutional Networks
• Subdiffusion and weak ergodicity breaking in the presence of a reactive boundary
• Improved estimation of anomalous diffusion exponents in single particle tracking experiments
• Measurement of Anomalous Diffusion Using Recurrent Neural Networks
• An introduction to infinite HMMs for single molecule data analysis
• On the Bottleneck of Graph Neural Networks and its Practical Implications
• Construction of and efficient sampling from the simplicial configuration model
• Learning force fields from stochastic trajectories
• Certain uncertainty: using pointwise error estimates in super-resolution microscopy
• Time-averaged MSD for switching diffusion
• First passage statistics for aging diffusion in annealed and quenched disorder
• Single-trajectory spectral analysis of scaled Brownian motion
• Unravelling intermittent features in single particle trajectories by a local convex hull method
• Distribution of the Time Between Maximum and Minimum of Random Walks
• Systems-level approach to uncovering diffusive states and their transitions from single particle trajectories
• Large deviations for continuous time random walks
• Optimal parameters for anomalous diffusion exponent estimation from noisy data
• Mean perimeter and mean area of the convex hull over planar random walks
• Motion Blur Filtering: A Statistical Approach for Extracting Confinement Forces and Diffusivity from a Single Blurred Trajectory
• AnDi: The Anomalous Diffusion Challenge
• Survival probability of a Brownian motion in a planar wedge of arbitrary angle
• Distributional Behavior of Diffusion Coefficients Obtained by Single Trajectories in Annealed Transit Time Model
• Modelling Point Spread Function in Fluorescence Microscopy with a Sparse Combination of Gaussian Mixture: Trade-off between Accuracy and Efficiency
• Record statistics for random walk bridges

Cited by in corpus (15)

• Bayesian deep learning for error estimation in the analysis of anomalous diffusion
• Geometric deep learning reveals the spatiotemporal fingerprint of microscopic motion
• Machine-Learning Solutions for the Analysis of Single-Particle Diffusion Trajectories
• Inferring pointwise diffusion properties of single trajectories with deep learning
• WaveNet-Based Deep Neural Networks for the Characterization of Anomalous Diffusion (WADNet)
• Semantic Segmentation of Anomalous Diffusion Using Deep Convolutional Networks
• Unsupervised learning of anomalous diffusion data
• Variational inference of fractional Brownian motion with linear computational complexity
• Temperature and friction fluctuations inside a harmonic potential
• Machine Learning Analysis of Anomalous Diffusion
• Objective comparison of methods to decode anomalous diffusion
• Preface: Characterisation of Physical Processes from Anomalous Diffusion Data
• Change-point detection in anomalous-diffusion trajectories utilising machine-learning-based uncertainty estimates
• Bottom-up Iterative Anomalous Diffusion Detector (BI-ADD)
• Recurrent neural network analysis of single trajectories switching between anomalous diffusion states