Short Trajectory Segmentation with 1D UNET Framework: Application to Secretory Vesicle Dynamics

Dmitrieva, Mariia; Lefebvre, Joël; Peñas, Kristofer delas; Zenner, Helen; Richens, Jennifer H.; Johnston, Daniel St; Rittscher, Jens

doi:10.1109/isbi45749.2020.9098426

Cited by 5 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results add to existing research on using U-Nets in 1-dimensional signal processing. Examples are the segmentation of electrocardiograms [63][64][65] , segmentation of vesicle trajectories 66 , and Raman and infrared spectra refinement. 67,68 To our knowledge, this is the first application of U-Nets on FCS intensity timeseries.…”

Section: Resultsmentioning

confidence: 99%

Neural Network Informed Photon Filtering Reduces Artifacts in Fluorescence Correlation Spectroscopy Data

Seltmann,

Carravilla,

Reglinski

et al. 2023

Preprint

View full text Add to dashboard Cite

Fluorescence Correlation Spectroscopy (FCS) techniques are well-established tools to investigate molecular dynamics in confocal and super-resolution microscopy. In practice, users often need to handle a variety of sample or hardware-related artifacts, an example being peak artifacts created by bright, slow-moving clusters. Approaches to address peak artifacts exist, but measurements suffering from severe artifacts are typically non-analyzable. Here, we trained a 1-dimensional U-Net to automatically identify peak artifacts in fluorescence time-series and then analyzed the purified, non-artifactual fluctuations by time-series editing. We show that in samples with peak artifacts, the transit time and particle number distributions can be restored in simulations and validated the approach in two independent biological experiments. We propose that it is adaptable for other FCS artifacts, such as detector dropout, membrane movement, or photobleaching. In conclusion, this simulation-based, automated, open-source pipeline makes measurements analyzable which previously had to be discarded and extends every FCS user's experimental toolbox.

show abstract

Section: Resultsmentioning

confidence: 99%

Neural Network Informed Photon Filtering Reduces Artifacts in Fluorescence Correlation Spectroscopy Data

Seltmann,

Carravilla,

Reglinski

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The speed is evaluated for the entire trajectory (average speed), for the trajectory segments where the particle is moving (moving speed), and the maximum curvilinear speed is calculated for a given time interval by a sliding window over the moving segments. To allow segmentation of short trajectories, where the traditional Mean Square Displacement (MSD) methods are not applicable, we exploit a 1D U-Net segmentation [75]. The method provides segmentation of the directed motion of the particle from the remaining trajectory.…”

Section: Methodsmentioning

confidence: 99%

Tracking exocytic vesicle movements reveals the spatial control of secretion in epithelial cells

Richens,

Dmitrieva,

Zenner

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Understanding how specific secretory cargoes are targeted to distinct domains of the plasma membrane in epithelial cells requires analysing the trafficking of post-Golgi vesicles to their sites of secretion. Here, we introduce an interactive vesicle tracking framework, the multiscale particle tracker and viewer, that exploits developments in computer vision and deep learning to determine vesicle trajectories in noisy cellular environments without the need for extensive training data. We analysed the movements of secreted cargoes in the Drosophila follicular epithelium ex vivo with high temporal and spatial resolution and found that MSP-tracker outperformed other tracking software. Using the RUSH (retention using selective hooks) system to synchronously release cargoes from the endoplasmic reticulum, we followed the movements the microvillar protein, Cadherin 99C, to the plasma membrane. A dominant slow dynein mutant reduces the speed of apical Cad99C movements, demonstrating that dynein transports Cad99C vesicles to the apical cortex. Furthermore, both the dynein mutant and microtubule depolymerisation cause lateral Cad99C secretion. Thus, microtubule organisation plays a central role in targeting secretion, suggesting that Drosophila does not have distinct apical versus basolateral vesicle fusion machinery. Imaging the extracellular matrix protein (ECM) proteins, Nidogen and Collagen IV revealed that Nidogen vesicles undergo planar-polarised transport to the leading edge of follicle cells as they migrate over the ECM, whereas most Collagen is secreted at trailing edges. The follicle cells therefore secrete different ECM components at opposite sides of the cell, revealing that the secretory pathway is more spatially organised than previously thought.

show abstract

“…For example, Stoller et al proved that a 1D U-net can be used to perform the audio source separation to achieve speech enhancement . Dmitrieva et al applied a 1D U-net to segment a given trajectory and explore the quantification of vesicle dynamics …”

Section: Introductionmentioning

confidence: 99%

“…22 Dmitrieva et al applied a 1D Unet to segment a given trajectory and explore the quantification of vesicle dynamics. 23 Thus, in this study, a 1D U-net network was applied to investigate the feasibility of LF-NMR spectrum calibration transfer. The calibration-transfer ability of the 1D U-net was evaluated by the spectral response, feature analysis (principal component analysis, PCA), and quantitative concentration analysis (support vector regression) on two real data (edible oils and cupric sulfate).…”

Section: ■ Introductionmentioning

confidence: 99%

A One-Dimensional U-Net-Based Calibration-Transfer Method for Low-Field Nuclear Magnetic Resonance Signals

et al. 2021

View full text Add to dashboard Cite

The reconstruction of the statistical analysis model of an instrument is a time-consuming and expensive process. Herein, the feasibility of spectral model calibration-transfer application to the same type of low-field nuclear magnetic resonance (LF-NMR) instrument was investigated using a one-dimensional U-net (1D U-net). Unlike conventional calibration-transfer algorithms such as direct standardization (DS), the 1D U-net network can reduce the error between the master and slave instruments through iterative cycles. The calibration-transfer ability was verified; three experiments that entailed the use of edible oil and copper sulfate (CuSO 4 ) samples were implemented. The analysis of the spectral responses and feature analysis of the edible oil samples revealed that the signal of the slave instrument calibrated using the 1D U-net most resembled the signal of the master instrument, and its relative residual value was reduced to 0.0045. Further analysis of the CuSO 4 concentration prediction showed that on the support vector regression (SVR) model constructed using the master instrument, the signal of the slave instrument calibrated by the 1D U-net was more similar to the response of the master instrument, and its root mean square error (RMSE) was only 0.01606 mmol/L. Thus, 1D U-net is a viable candidate for calibration-transfer applications to LF-NMR instruments.

show abstract

Short Trajectory Segmentation with 1D UNET Framework: Application to Secretory Vesicle Dynamics

Cited by 5 publications

References 11 publications

Neural Network Informed Photon Filtering Reduces Artifacts in Fluorescence Correlation Spectroscopy Data

Neural Network Informed Photon Filtering Reduces Artifacts in Fluorescence Correlation Spectroscopy Data

Tracking exocytic vesicle movements reveals the spatial control of secretion in epithelial cells

A One-Dimensional U-Net-Based Calibration-Transfer Method for Low-Field Nuclear Magnetic Resonance Signals

Contact Info

Product

Resources

About