Article
Mathematics
Dwueng-Chwuan Jhwueng
Summary: Gaussian processes, specifically the Brownian bridge and Ornstein-Uhlenbeck bridge, are proposed to model continuous trait evolution of related species along phylogenetic trees. Traitgrams are generated to display the evolutionary trajectories. These novel models are applied to study body mass evolution of a group of marsupial species.
Article
Mathematics, Applied
Massimiliano Tamborrino, Petr Lansky
Summary: The study focuses on nonnegative shot noise processes and their weak convergence to Levy-driven Ornstein-Uhlenbeck (OU) process, highlighting the importance of underlying jump distributions in determining the features. It identifies necessary conditions for diffusion limit to a Gaussian OU process, indicating the need to consider negative jumps happening with probability approaching zero.
PHYSICA D-NONLINEAR PHENOMENA
(2021)
Article
Computer Science, Artificial Intelligence
Tao Li, Jinwen Ma
Summary: In this paper, a Dirichlet Process Mixture of Gaussian Process Functional Regressions (DPM-GPFR) is proposed to address the heterogeneity problem in functional data analysis and achieve flexible model fitting. Experimental results demonstrate the effectiveness of the proposed method.
PATTERN RECOGNITION
(2022)
Article
Mathematics
Pece Trajanovski, Petar Jolakoski, Ljupco Kocarev, Trifce Sandev
Summary: This study investigates anomalous transport in a three-dimensional comb structure using the Ornstein-Uhlenbeck (O-U) process with resetting. The three-dimensional comb structure consists of backbones and fingers, with x-axis corresponding to the backbones and y-axis and z-axis corresponding to the fingers. Implementation of the O-U process on the comb structure leads to anomalous (non-Markovian) diffusion. This specific anomalous transport with resetting results in non-equilibrium stationary states. Analytical expressions for the mean values and mean squared displacements along all three directions of the comb are obtained and numerically verified. The marginal probability density functions for each direction are obtained through Monte Carlo simulation of a random transport described by a system of coupled Langevin equations for the comb geometry.
Article
Automation & Control Systems
Peter Koepernik, Florian Pfaff
Summary: This paper investigates the formal consistency of Gaussian process (GP) regression on general metric spaces, demonstrating that the variance of the posterior GP converges to zero almost surely and the posterior mean converges pointwise in L-2 to the unknown function.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Biochemical Research Methods
Pierre Barrat-Charlaix, Timothy G. Vaughan, Richard A. Neher
Summary: When two influenza viruses co-infect the same cell, they can exchange genome segments through reassortment, which is an important source of genetic diversity and has been involved in the emergence of most pandemic influenza strains. A new method called TreeKnit is introduced to infer ancestral reassortment graphs (ARG) from two segment trees. The performance of TreeKnit is demonstrated to be as accurate as a more principled bayesian method but much faster. Additionally, it is shown that the inferred ARG can be used to improve the resolution of segment trees and construct more informative visualizations of reassortments.
PLOS COMPUTATIONAL BIOLOGY
(2022)
Article
Mathematics
Fehaid Salem Alshammari, Fahir Talay Akyildiz
Summary: This report solves the problem of disease behavior in the stochastic SIRVI model and proposes a new model. Through mathematical derivation and algorithm calculation, we obtain the global solution and extinction conditions for the disease transmission rate. Additionally, we present numerical solutions using COVID-19 data.
Article
Physics, Multidisciplinary
Noa Malem-Shinitski, Cesar Ojeda, Manfred Opper
Summary: In this paper, we propose an extended model to simulate time-continuous point processes with history dependence. The self-effects in our model can be both excitatory and inhibitory, following a Gaussian Process. Compared to previous methods, our formulation allows for flexible model and learning even when data is scarce.
Article
Multidisciplinary Sciences
Felipe Tobar, Arnaud Robert, Jorge F. F. Silva
Summary: This article discusses the problem of blind deconvolution and proposes a novel strategy for Bayesian non-parametric deconvolution using a Gaussian process prior when x is a continuous-time signal. The conditions for a well-defined direct model are analyzed, and the feasibility of Bayesian deconvolution and the learning or approximation of the filter h for blind deconvolution are studied. The proposed Gaussian process deconvolution approach is compared to other methods conceptually and validated using examples and real-world datasets.
PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
(2023)
Article
Biochemistry & Molecular Biology
Devin P. Bendixsen, Tanner B. Pollock, Gianluca Peri, Eric J. Hayden
Summary: The study used a phylogenetic approach to evaluate the activity changes of self-cleaving ribozymes during mammalian evolution, finding that high activity ribozymes have been conserved in most lineages. Additionally, a reduction in ribozyme activity was observed multiple times throughout mammalian evolution.
MOLECULAR BIOLOGY AND EVOLUTION
(2021)
Review
Engineering, Mechanical
Xiaoxu Wang, Haoran Cui, Tiancheng Li, Yan Liang, Zhengtao Ding
Summary: A new variational Gaussian regression filter is proposed in this paper by incorporating variational parameters into a linear parametric Gaussian regression process. The filtering evidence lower bound serves as a quantitative evaluation rule for different filters, and a relationship between F-ELBO and M-ELBO is identified. The accuracy performance improvement of VGRF can be theoretically explained based on these findings.
NONLINEAR DYNAMICS
(2021)
Article
Mathematics, Applied
Yaxin Zhou, Daqing Jiang
Summary: This study considers the dynamical behaviors of a stochastic SIQR epidemic model with mean-reverting Ornstein-Uhlenbeck process and standard incidence under the continuous interference of environmental white noise. After dimensionality reduction, several conclusions are derived, including the existence and uniqueness of positive solution, a sufficient condition for extinction of the diseases, and the stationary distribution of the model. Furthermore, an exact local expression of the density function of the random model near the unique endemic equilibrium is proposed, and numerical simulations are performed to validate the theoretical results.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2023)
Article
Engineering, Mechanical
Osmo Kaleva, Heikki Orelma
Summary: This paper is a continuation of the study on the continuum approach to high-cycle fatigue model. It models stress history as a stochastic process to simulate fatigue evolution and lifetime distribution.
INTERNATIONAL JOURNAL OF FATIGUE
(2021)
Article
Mathematics, Applied
Qun Liu
Summary: In this paper, a stochastic SISP respiratory disease system with Ornstein-Uhlenbeck process is established and analyzed to describe the transmission dynamics of respiratory disease in the population. The existence and uniqueness of a global solution for any initial value are verified. Sufficient criteria for the existence of a stationary distribution of positive solutions are obtained using a stochastic Lyapunov function method.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2023)
Article
Mathematics, Interdisciplinary Applications
Miaomiao Gao, Daqing Jiang, Jieyu Ding
Summary: In this paper, the dynamics of a nutrient-phytoplankton-zooplankton model with nutrient recycling is investigated, where the maximal nutrient uptake rate and maximal zooplankton ingestion rate are given by a continuous, mean-reverting, stochastic process. The existence and uniqueness of the global solution is first proven. Conditions for the extinction of plankton are then derived in two cases. Furthermore, by constructing appropriate Lyapunov functions, sufficient condition for the existence of stationary distribution is established. It is worth noting that the exact expression of density function around the positive equilibrium of deterministic system is also provided. Finally, simulations are conducted to demonstrate the theoretical results.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Multidisciplinary Sciences
Kyunghee Han, Pantelis Z. Hadjipantelis, Jane-Ling Wang, Michael S. Kramer, Seungmi Yang, Richard M. Martin, Hans-Georg Mueller
Article
Neurosciences
Xiongtao Dai, Pantelis Hadjipantelis, Jane-Ling Wang, Sean C. L. Deoni, Hans-Georg Muller
HUMAN BRAIN MAPPING
(2019)
Article
Immunology
Gurdeep Singh, Andrew Brass, Christopher G. Knight, Sheena M. Cruickshank
Article
Multidisciplinary Sciences
Rok Krasovec, Huw Richards, Guillaume Gomez, Danna R. Gifford, Adrien Mazoyer, Christopher G. Knight
JOVE-JOURNAL OF VISUALIZED EXPERIMENTS
(2019)
Article
Evolutionary Biology
Joseph N. Keating, Robert S. Sansom, Mark D. Sutton, Christopher G. Knight, Russell J. Garwood
SYSTEMATIC BIOLOGY
(2020)
Review
Multidisciplinary Sciences
Franciska T. de Vries, Rob I. Griffiths, Christopher G. Knight, Oceane Nicolitch, Alex Williams
Review
Immunology
Paul M. Campbell, Gavin J. Humphreys, Angela M. Summers, Joanne E. Konkel, Christopher G. Knight, Titus Augustine, Andrew J. McBain
FRONTIERS IN CELLULAR AND INFECTION MICROBIOLOGY
(2020)
Article
Multidisciplinary Sciences
Gurdeep Singh, Andrew Brass, Sheena M. Cruickshank, Christopher G. Knight
Summary: Findings from a mouse gut microbiome experiment show that host social groups, age, and niche significantly impact the microbial community structure. This study suggests the importance of considering environmental factors in microbiome research to understand how gut microbial communities are affected by various ecological and experimental factors.
SCIENTIFIC REPORTS
(2021)
Article
Multidisciplinary Sciences
Pierluigi Mancarella, John Moriarty, Andy Philpott, Almut Veraart, Stan Zachary, Bert Zwart
Summary: The urgent need to decarbonize energy systems has led to interdisciplinary research involving mathematicians, physicists, engineers, and economists. Renewable energy generation, particularly wind and solar, presents challenges in variability and predictability, as well as issues in controlling and optimizing power and energy systems for market liberalization. Planning and investment in energy systems also pose physical and economic design challenges on longer time scales.
PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
(2021)
Article
Engineering, Electrical & Electronic
Jhonny Gonzalez, Panagiotis N. Papadopoulos, Jovica V. Milanovic, Goran Peskir, John Moriarty
Summary: This paper addresses the problem of 'quickest possible' online transient stability assessment by optimizing event detection and generator group prediction while respecting predefined probabilistic error constraints. The approach is shown to be two to three times faster on average than strategies based on fixed assessment times with comparable error rates in simulated measurements from interconnected systems.
IEEE TRANSACTIONS ON SMART GRID
(2021)
Article
Biology
Sam Clark, Thomas A. Jowitt, Lynda K. Harris, Christopher G. Knight, Curtis B. Dobson
Summary: Clark et al. comprehensively explore the primary structural features underlying the activity of a complete set of antimicrobial peptides (AMPs). They find that the shortest active peptides were 4 or 5 residues in length, with activity being associated with 40% arginine, and multiple adjacent tryptophan residues. This study provides insights into the design of effective AMPs.
COMMUNICATIONS BIOLOGY
(2021)
Article
Statistics & Probability
Randall Martyr, John Moriarty, Magnus Perninge
Summary: We address the non-Markovian optimal switching problems in discrete time on an infinite horizon, considering the risk-awareness of the decision-maker and the general filtration. We establish the existence and uniqueness of solutions for the associated reflected backward stochastic difference equations and provide an example application to hydropower planning.
ADVANCES IN APPLIED PROBABILITY
(2022)
Article
Operations Research & Management Science
Tomasz Kosmala, Randall Martyr, John Moriarty
Summary: In this paper, a probabilistic Markov property is formulated in discrete time under a dynamic risk framework. This property is useful for recursive solutions to risk-sensitive versions of dynamic optimization problems, such as optimal prediction. The property holds for standard measures of risk used in practice and can be formulated in several equivalent versions.
MATHEMATICAL METHODS OF OPERATIONS RESEARCH
(2023)
Article
Genetics & Heredity
Danna R. Gifford, Ernesto Berrios-Caro, Christine Joerres, Marc Sune, Jessica H. Forsyth, Anish Bhattacharyya, Tobias Galla, Christopher G. Knight
Summary: Combination therapy of antibiotics aims to prevent the evolution of resistance, but researchers have found that mutators with defects in DNA repair can readily evolve resistance in both single-drug and combination treatments. This suggests that combination therapy may not be as effective against resistance evolution as once thought.
Article
Business, Finance
John Moriarty, Jan Palczewski
