Estimation of diffusive states from single-particle trajectory in heterogeneous medium using machine-learning methods

Matsuda, Yu; Hanasaki, Itsuo; Iwao, Ryo; Yamaguchi, Hiroki; Niimi, Tomohide

doi:10.1039/c8cp02566e

Cited by 21 publications

(14 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The time-dependent squared cell velocity averaged over all cells is shown in Fig. 1(b) and turns out to be rather constant for times larger than 2 h. This indicates that cell motion is stationary for long times (nonstationary cell dynamics has been considered in previous works [38,39]). We therefore discard the data for the first two hours (shown in red) for all further analysis (see Appendix B for further details).…”

Section: A Experimental Trajectoriesmentioning

confidence: 71%

Non-Markovian data-driven modeling of single-cell motility

et al. 2020

View full text Add to dashboard Cite

Trajectories of human breast cancer cells moving on one-dimensional circular tracks are modeled by the non-Markovian version of the Langevin equation that includes an arbitrary memory function. When averaged over cells, the velocity distribution exhibits spurious non-Gaussian behavior, while single cells are characterized by Gaussian velocity distributions. Accordingly, the data are described by a linear memory model which includes different random walk models that were previously used to account for various aspects of cell motility such as migratory persistence, non-Markovian effects, colored noise, and anomalous diffusion. The memory function is extracted from the trajectory data without restrictions or assumptions, thus making our approach truly data driven, and is used for unbiased single-cell comparison. The cell memory displays time-delayed single-exponential negative friction, which clearly distinguishes cell motion from the simple persistent random walk model and suggests a regulatory feedback mechanism that controls cell migration. Based on the extracted memory function we formulate a generalized exactly solvable cell migration model which indicates that negative friction generates cell persistence over long timescales. The nonequilibrium character of cell motion is investigated by mapping the non-Markovian Langevin equation with memory onto a Markovian model that involves a hidden degree of freedom and is equivalent to the underdamped active Ornstein-Uhlenbeck process.

show abstract

Section: A Experimental Trajectoriesmentioning

confidence: 71%

Non-Markovian data-driven modeling of single-cell motility

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Although examining individual trajectories and developing novel anomalous diffusion models may provide new insights into the molecular events occurring in the plasma membrane ( Dietrich et al , 2002 ; Fujiwara et al , 2016 ; Ghosh et al , 2019 ; Jin et al , 2007 ; Liu et al , 2016 ; Zhao et al , 2019 ), we believe we can also extract hidden features in the membrane from a vast amount of receptor trajectories using deep learning algorithms. Unlike the previous reports that focused on diffusive state characterization using the trajectory-trained machine learning or deep learning models ( Dosset et al , 2016 ; Granik et al , 2019 ; Matsuda et al , 2018 ; Wagner et al , 2017 ), here we directly differentiated cell types based on hidden features extracted from the transmembrane receptor trajectories.…”

Section: Introductionmentioning

confidence: 99%

Deep learning-based classification of breast cancer cells using transmembrane receptor dynamics

et al. 2021

View full text Add to dashboard Cite

Motivation Motions of transmembrane receptors on cancer cell surfaces can reveal biophysical features of the cancer cells, thus providing a method for characterizing cancer cell phenotypes. While conventional analysis of receptor motions in the cell membrane mostly relies on the mean-squared displacement plots, much information is lost when producing these plots from the trajectories. Here we employ deep learning to classify breast cancer cell types based on the trajectories of epidermal growth factor receptor (EGFR). Our model is an artificial neural network trained on the EGFR motions acquired from six breast cancer cell lines of varying invasiveness and receptor status: MCF7 (hormone receptor-positive), BT474 (HER2-positive), SKBR3 (HER2-positive), MDA-MB-468 (triple-negative, TN), MDA-MB-231 (TN), and BT549 (TN). Results The model successfully classified the trajectories within individual cell lines with 83% accuracy and predicted receptor status with 85% accuracy. To test the capability of the trained neural network, epithelial-mesenchymal transition (EMT) was induced in benign MCF10A cells, non-invasive MCF7 cancer cells and highly invasive MDA-MB-231 cancer cells, and EGFR trajectories from these cells were tested. As expected, after EMT induction, both MCF10A and MCF7 cells showed higher rates of classification as TN cells but not the MDA-MB-231 cells. Whereas deep learning-based cancer cell classifications are primarily based on the optical transmission images of cell morphology or the fluorescence images of cell organelle or cytoskeleton structures, here we demonstrated an alternative way to classify cancer cells using a dynamic, biophysical feature that is readily accessible. Availability A python implementation of deep learning-based classification can be found at https://github.com/soonwoohong/Deep-learning-for-EGFR-trajectory-classification. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

“…An essential strength of SPT methods is that the list of particle positions acquired during a measurement contains temporal information. This feature can be exploited to identify transient periods of statistically similar motion within the same trajectory, including different diffusion states (7)(8)(9), changes in diffusion type, e.g. distinguishing between Brownian, confined and directed diffusion (10)(11)(12) and the associated kinetics of transitions and equilibrium probabilities.…”

Section: Introductionmentioning

confidence: 99%

“…This makes the problem of single-particle diffusion characterization well suited for deep-learning analysis. Promisingly, several groups have applied classical machine learning (ML) (7) and deep learning (11,12) algorithms to classify diffusion trajectories to confined, directed and normal diffusion, showing some advantage over traditional methods. For example, Muñoz-Gil and co-workers (22) recently used a Random Forest algorithm to classify a given trajectory as one of several anomalous diffusion models, addressing a similar problem.…”

Section: Introductionmentioning

confidence: 99%

Single particle diffusion characterization by deep learning

Granik

Nehme

Weiss

et al. 2019

Preprint

View full text Add to dashboard Cite

Diffusion plays a crucial role in many biological processes including signaling, cellular organization, transport mechanisms, and more. Direct observation of molecular movement by single-particle tracking experiments has contributed to a growing body of evidence that many cellular systems do not exhibit classical Brownian motion, but rather anomalous diffusion. Despite this evidence, characterization of the physical process underlying anomalous diffusion remains a challenging problem for several reasons. First, different physical processes can exist simultaneously in a system. Second, commonly used tools to distinguish between these processes are based on asymptotic behavior, which is inaccessible experimentally in most cases. Finally, an accurate analysis of the diffusion model requires the calculation of many observables, since different transport modes can result in the same diffusion power-law α, that is obtained from the commonly used mean-squared-displacement (MSD) in its various forms. The outstanding challenge in the field is to develop a method to extract an accurate assessment of the diffusion process using many short trajectories with a simple scheme that is applicable at the non-expert level.Here, we use deep learning to infer the underlying process resulting in anomalous diffusion. We implement a neural network to classify single particle trajectories according to diffusion type -Brownian motion, fractional Brownian motion (FBM) and Continuous Time Random Walk (CTRW). We further use the net to estimate the Hurst exponent for FBM, and the diffusion coefficient for Brownian motion, demonstrating its applicability on simulated and experimental data. The networks outperform time averaged MSD analysis on simulated trajectories while requiring as few as 25 time-steps. Furthermore, when tested on experimental data, both network and ensemble MSD analysis converge to similar values, with the network requiring half the trajectories required for ensemble MSD. Finally, we use the nets to extract diffusion parameters from multiple extremely short trajectories (10 steps).

show abstract

Estimation of diffusive states from single-particle trajectory in heterogeneous medium using machine-learning methods

Abstract: We propose a novel approach to analyze random walks in heterogeneous medium using a hybrid machine-learning method based on a gamma mixture and a hidden Markov model.

Cited by 21 publications

References 44 publications

Non-Markovian data-driven modeling of single-cell motility

Non-Markovian data-driven modeling of single-cell motility

Deep learning-based classification of breast cancer cells using transmembrane receptor dynamics

Single particle diffusion characterization by deep learning

Contact Info

Product

Resources

About