New efficient algorithms for multiple change-point detection with reproducing kernels

Uncovering loggerhead (Caretta caretta) navigation strategy in the open ocean through the consideration of their diving behaviour

While scientists have been monitoring the movements and diving behaviour of sea turtles using Argos platform terminal transmitters for decades, the precise navigational mechanisms used by these animals remain an open question. Until now, active swimming motion has been derived from total motion by subtracting surface or subsurface modelled ocean currents, following the approximation of a quasi-two-dimensional surface layer migration. This study, based on tracking and diving data collected from 25 late-juvenile loggerhead turtles released from Reunion Island during their pre-reproductive migration, demonstrates the importance of considering the subsurface presence of the animals. Using a piecewise constant heading model, we investigate navigation strategy using daily time-at-depth distributions and three-dimensional currents to calculate swimming velocity. Our results are consistent with a map and compass strategy in which swimming movements follow straight courses at a stable swimming speed (approx. 0.5 m s-1), intermittently segmented by course corrections. This strategy, previously hypothesized for post-nesting green and hawksbill turtles, had never been observed in juvenile loggerheads. These results confirm a common open-ocean navigation mechanism across ages and species and highlight the importance of considering diving behaviour in most studies of sea turtle spatial ecology.

Inferring pointwise diffusion properties of single trajectories with deep learning

To characterize the mechanisms governing the diffusion of particles in biological scenarios, it is essential to accurately determine their diffusive properties. To do so, we propose a machine-learning method to characterize diffusion processes with time-dependent properties at the experimental time resolution. Our approach operates at the single-trajectory level predicting the properties of interest, such as the diffusion coefficient or the anomalous diffusion exponent, at every time step of the trajectory. In this way, changes in the diffusive properties occurring along the trajectory emerge naturally in the prediction and thus allow the characterization without any prior knowledge or assumption about the system. We first benchmark the method on synthetic trajectories simulated under several conditions. We show that our approach can successfully characterize both abrupt and continuous changes in the diffusion coefficient or the anomalous diffusion exponent. Finally, we leverage the method to analyze experiments of single-molecule diffusion of two membrane proteins in living cells: the pathogen-recognition receptor DC-SIGN and the integrin α5β1. The analysis allows us to characterize physical parameters and diffusive states with unprecedented accuracy, shedding new light on the underlying mechanisms.

Automated contour extraction for light-sheet microscopy images of zebrafish embryos based on object edge detection algorithm

Embryo contour extraction is the initial step in the quantitative analysis of embryo morphology, and it is essential for understanding the developmental process. Recent developments in light-sheet microscopy have enabled the in toto time-lapse imaging of embryos, including zebrafish. However, embryo contour extraction from images generated via light-sheet microscopy is challenging owing to the large amount of data and the variable sizes, shapes, and textures of objects. In this report, we provide a workflow for extracting the contours of zebrafish blastula and gastrula without contour labeling of an embryo. This workflow is based on the edge detection method using a change point detection approach. We assessed the performance of the edge detection method and compared it with widely used edge detection and segmentation methods. The results showed that the edge detection accuracy of the proposed method was superior to those of the Sobel, Laplacian of Gaussian, adaptive threshold, Multi Otsu, and k-means clustering-based methods, and the noise robustness of the proposed method was superior to those of the Multi Otsu and k-means clustering-based methods. The proposed workflow was shown to be useful for automating small-scale contour extractions of zebrafish embryos that cannot be specifically labeled owing to constraints, such as the availability of microscopic channels. This workflow may offer an option for contour extraction when deep learning-based approaches or existing non-deep learning-based methods cannot be applied.

Railway Axle Early Fatigue Crack Detection through Condition Monitoring Techniques

The detection of cracks in rotating machinery is an unresolved issue today. In this work, a methodology for condition monitoring of railway axles is presented, based on crack detection by means of the automatic selection of patterns from the vibration signal measurement. The time waveforms were processed using the Wavelet Packet Transform, and appropriate alarm values for diagnosis were calculated automatically using non-supervised learning techniques based on Change Point Analysis algorithms. The validation was performed using vibration signals obtained during fatigue tests of two identical railway axle specimens, one of which cracked during the test while the other did not. During the test in which the axle cracked, the results show trend changes in the energy of the vibration signal associated with theoretical defect frequencies, which were particularly evident in the direction of vibration that was parallel to the track. These results are contrasted with those obtained during the test in which the fatigue limit was not exceeded, and the test therefore ended with the axle intact, verifying that the effects that were related to the crack did not appear in this case. With the results obtained, an adjusted alarm value for a condition monitoring process was established.

Unfolding and modeling the recovery process after COVID lockdowns

Lockdown is a common policy used to deter the spread of COVID-19. However, the question of how our society comes back to life after a lockdown remains an open one. Understanding how cities bounce back from lockdown is critical for promoting the global economy and preparing for future pandemics. Here, we propose a novel computational method based on electricity data to study the recovery process, and conduct a case study on the city of Hangzhou. With the designed Recovery Index, we find a variety of recovery patterns in main sectors. One of the main reasons for this difference is policy; therefore, we aim to answer the question of how policies can best facilitate the recovery of society. We first analyze how policy affects sectors and employ a change-point detection algorithm to provide a non-subjective approach to policy assessment. Furthermore, we design a model that can predict future recovery, allowing policies to be adjusted accordingly in advance. Specifically, we develop a deep neural network, TPG, to model recovery trends, which utilizes the graph structure learning to perceive influences between sectors. Simulation experiments using our model offer insights for policy-making: the government should prioritize supporting sectors that have greater influence on others and are influential on the whole economy.

A novel machine learning-based approach for the detection and analysis of spontaneous synaptic currents

Spontaneous synaptic activity is a hallmark of biological neural networks. A thorough description of these synaptic signals is essential for understanding neurotransmitter release and the generation of a postsynaptic response. However, the complexity of synaptic current trajectories has either precluded an in-depth analysis or it has forced human observers to resort to manual or semi-automated approaches based on subjective amplitude and area threshold settings. Both procedures are time-consuming, error-prone and likely affected by human bias. Here, we present three complimentary methods for a fully automated analysis of spontaneous excitatory postsynaptic currents measured in major cell types of the mouse retina and in a primary culture of mouse auditory cortex. Two approaches rely on classical threshold methods, while the third represents a novel machine learning-based algorithm. Comparison with frequently used existing methods demonstrates the suitability of our algorithms for an unbiased and efficient analysis of synaptic signals in the central nervous system.

Development of an Objective Low Flow Identification Method Using Breakpoint Analysis

Low flow events (a.k.a. streamflow drought) are described as episodes where stream flows are lower or equal to a specified minimum threshold level. This threshold is usually predefined at the methodological stage of a study and is generally applied as a chosen flow percentile, determined from a flow duration curve (FDC). Unfortunately, many available methods for choosing both the percentile and FDCs result in a large range of potential thresholds, which reduces the ability to statistically compare the results from the different methods while also losing the natural character of the phenomenon. The aim of this work is to introduce a new approach for low flow threshold calculation through the application of an objective approach using breakpoint analysis. This method allows for the identification of an environmental moment of river transition, from atmospheric feed flows to base flow, which characterizes the moment at the beginning of the hydrological drought. The method allows for not only the capture of the genesis of a low flow event but, above all, unifies the approach toward threshold levels and completely excludes the impact of the subjective researcher’s decisions, which occur at the methodological stage when selecting the threshold criteria or when choosing a respective percentile. In addition, the method can be successfully used in datasets characterized by a high level of discretization, such as numerical model data, where the subsurface runoff component is not described in sufficient detail. Results of this work show that the objective identification method is better able to capture the occurrence of a low flow event, improving the ability to identify hydrologic drought conditions. The proposed method is published together with the Python module objective_thresholds for broad use in other studies.

A Novel Online Change Point Detection Using an Approximate Random Blanket and the Line Process Energy

In this paper, we focus on the problem of change point detection in piecewise constant signals. This problem is central to several applications such as human activity analysis, speech or image analysis and anomaly detection in genetics. We present a novel window-sliding algorithm for an online change point detection. The proposed approach considers a local blanket of a global Markov Random Field (MRF) representing the signal and its noisy observation. For each window, we define and solve the local energy minimization problem to deduce the gradient on each edge of the MRF graph. The gradient is then processed by an activation function to filter the weak features and produce the final jumps. We demonstrate the effectiveness of our method by comparing its running time and several detection metrics with state of the art algorithms.

Segmentation Method for Ship-Radiated Noise Using the Generalized Likelihood Ratio Test on an Ordinal Pattern Distribution

Due to the diversity of ship-radiated noise (SRN), audio segmentation is an essential procedure in the ship statuses/categories identification. However, the existing segmentation methods are not suitable for the SRN because of the lack of prior knowledge. In this paper, by a generalized likelihood ratio (GLR) test on the ordinal pattern distribution (OPD), we proposed a segmentation criterion and introduce it into single change-point detection (SCPD) and multiple change-points detection (MCPD) for SRN. The proposed method is free from the acoustic feature extraction and the corresponding probability distribution estimation. In addition, according to the sequential structure of ordinal patterns, the OPD is efficiently estimated on a series of analysis windows. By comparison with the Bayesian Information Criterion (BIC) based segmentation method, we evaluate the performance of the proposed method on both synthetic signals and real-world SRN. The segmentation results on synthetic signals show that the proposed method estimates the number and location of the change-points more accurately. The classification results on real-world SRN show that our method obtains more distinguishable segments, which verifies its effectiveness in SRN segmentation.

Adjacency-constrained hierarchical clustering of a band similarity matrix with application to genomics

BACKGROUND

Genomic data analyses such as Genome-Wide Association Studies (GWAS) or Hi-C studies are often faced with the problem of partitioning chromosomes into successive regions based on a similarity matrix of high-resolution, locus-level measurements. An intuitive way of doing this is to perform a modified Hierarchical Agglomerative Clustering (HAC), where only adjacent clusters (according to the ordering of positions within a chromosome) are allowed to be merged. But a major practical drawback of this method is its quadratic time and space complexity in the number of loci, which is typically of the order of 10 4 to 10 5 for each chromosome.

RESULTS

By assuming that the similarity between physically distant objects is negligible, we are able to propose an implementation of adjacency-constrained HAC with quasi-linear complexity. This is achieved by pre-calculating specific sums of similarities, and storing candidate fusions in a min-heap. Our illustrations on GWAS and Hi-C datasets demonstrate the relevance of this assumption, and show that this method highlights biologically meaningful signals. Thanks to its small time and memory footprint, the method can be run on a standard laptop in minutes or even seconds.

AVAILABILITY AND IMPLEMENTATION

Software and sample data are available as an R package, adjclust, that can be downloaded from the Comprehensive R Archive Network (CRAN).