Article
Computer Science, Artificial Intelligence
Jinhee Park, Junseok Kwon
Summary: In this study, novel visual tracking methods based on the Wasserstein approximate Bayesian computation (ABC) are presented. The proposed methods, including Wasserstein ABC (WABC), time-series WABC (TWABC), and Hilbert TWABC (HTWABC), improve the accuracy and efficiency of visual tracking by approximating likelihood distributions and encoding temporal dependencies. Experimental results demonstrate the superiority of the proposed methods, and ablation studies confirm the effectiveness of individual components.
PATTERN RECOGNITION
(2022)
Article
Biochemistry & Molecular Biology
Asher Moshe, Elya Wygoda, Noa Ecker, Gil Loewenthal, Oren Avram, Omer Israeli, Einat Hazkani-Covo, Itsik Pe'er, Tal Pupko
Summary: This study developed a probabilistic approach to infer genome rearrangement rate parameters and used an Approximate Bayesian Computation framework for inference. The method can help elucidate the role of genome rearrangement in evolution and simulate genomes with empirical dynamics.
MOLECULAR BIOLOGY AND EVOLUTION
(2022)
Article
Engineering, Industrial
Manuel Chiachio, Ali Saleh, Susannah Naybour, Juan Chiachio, John Andrews
Summary: This paper presents a probabilistic method for accurately defining the structure and parameters of a reduced Petri net model, enabling accurate modeling of engineering systems and processes. The method allows for numerical measurement and selection of the optimal simplified structure.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Civil
Pinghe Ni, Qiang Han, Xiuli Du, Xiaowei Cheng, Hongyuan Zhou
Summary: This paper presents a data-driven approach for post-earthquake reliability assessments of civil structures. It updates the probability density functions of random variables using measured vibration data, and generates the posterior probability density functions of structural parameters using two approximate Bayesian computation techniques. The updated probability density functions are then used for reliability assessments, and numerical studies verify the accuracy and efficiency of the proposed techniques.
ENGINEERING STRUCTURES
(2022)
Article
Computer Science, Interdisciplinary Applications
Clara Grazian, Luciana Dalla Valle, Brunero Liseo
Summary: Copula models are flexible tools for representing complex dependence structures of multivariate random variables. However, incorporating covariates in these models is challenging due to the lack of unbiased estimators of the conditional copula and the difficulty in selecting the appropriate copula model.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2022)
Article
Biochemistry & Molecular Biology
Cesar A. Fortes-Lima, Romain Laurent, Valentin Thouzeau, Bruno Toupance, Paul Verdu
Summary: Admixture is a fundamental evolutionary process that has influenced genetic patterns in numerous species. Maximum-likelihood approaches have limitations in reconstructing complex admixture histories, leading to the development of an Approximate Bayesian Computation (ABC) framework that integrates machine-learning algorithms to investigate admixture history accurately. The study showed that random forest ABC scenario-choice accurately distinguished complex admixture scenarios, while neural network ABC posterior parameter estimation was accurate and conservative under complex admixture scenarios. This approach provides a promising method for reconstructing detailed admixture histories in populations with multiple admixture pulses.
MOLECULAR ECOLOGY RESOURCES
(2021)
Article
Automation & Control Systems
Yuexi Wang, Tetsuya Kaji, Veronika Rockova
Summary: Approximate Bayesian Computation (ABC) is a method that enables statistical inference in simulator-based models with difficult likelihood calculations but easy simulation. This study constructs a kernel-type approximation of the posterior distribution in ABC by comparing summary statistics of real and simulated data, and uses contrastive learning to directly compare empirical distributions. The research demonstrates the usefulness of this approach in simulated examples and real data analysis in the context of stock volatility estimation.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Computer Science, Interdisciplinary Applications
Bach Do, Makoto Ohsaki
Summary: This study presents Bayesian optimization-assisted approximate Bayesian computation (BO-assisted ABC) and showcases its application to identifying the approximate posteriors of parameters for known statistical models and of cyclic elastoplastic parameters for structural steels. ABC bypasses likelihood evaluations by generating prior samples that are assigned as samples constituting the posterior if discrepancies between the experimental dataset and the corresponding simulated datasets do not exceed a small, positive threshold. With a modest number of costly simulations, BO facilitates ABC by intelligently constructing a Gaussian process model that approximates the discrepancy mean function.
COMPUTERS & STRUCTURES
(2023)
Article
Ecology
Charlotte Baey, Henrik G. Smith, Maj Rundlof, Ola Olsson, Yann Clough, Ullrika Sahlin
Summary: In this study, alternative strategies for handling high-dimensional data in ABC calibration were investigated. Regression adjustment and machine learning approaches were compared in terms of model estimate coverage and true parameter values. The results showed that random forest-based approaches performed better, and nonlinear regression adjustment outperformed linear ones.
ECOLOGICAL MODELLING
(2023)
Article
Engineering, Mechanical
Joao Pedro Valeriano, Pedro Henrique Cintra, Gustavo Libotte, Igor Reis, Felipe Fontinele, Renato Silva, Sandra Malta
Summary: The long duration of COVID-19 pandemic has led to multiple bursts in infection and death rates, known as epidemic waves. Traditional compartmental models are no longer effective, necessitating more sophisticated mathematical techniques for analyzing epidemic data and making reliable forecasts. In this study, a framework is proposed for analyzing complex dynamical systems by dividing the data into consecutive time-windows for separate analysis. Parameters are estimated for each window using an approximate Bayesian computation (ABC) algorithm, with the posterior distribution from one window serving as the prior distribution for the next. The Bayesian learning approach is tested using COVID-19 case data from various countries, showing improved ABC performance and producing accurate short-term forecasts.
NONLINEAR DYNAMICS
(2023)
Article
Physics, Multidisciplinary
Ilze A. Auzina, Jakub M. Tomczak
Summary: The study aims to address the lack of an optimal Approximate Bayesian Computation method for discrete random variables. The researchers propose an adjusted population-based MCMC ABC method and introduce a novel Markov kernel inspired by differential evolution to redefine ABC parameters as discrete ones. The results indicate the high potential of the proposed framework and the superiority of the new Markov kernel.
Article
Engineering, Civil
Zijie Zeng, Min Gao, Ching Tai Ng, Abdul Hamid Sheikh
Summary: This paper presents a probabilistic framework using ultrasonic guided waves to detect and identify early-state cracks in pipe-like structures. The framework quantifies the values and uncertainties of crack location, crack sizes, and Young's modulus through a Bayesian approach. The proposed framework utilizes an approximate Bayesian computation algorithm to estimate the posterior distributions of unknown parameters. The accuracy and practicability of the framework are validated through numerical and experimental case studies. Evaluation: 9/10
THIN-WALLED STRUCTURES
(2023)
Article
Mathematics, Interdisciplinary Applications
Evgeny Levi, Radu Craiu
Summary: Scientists utilize large datasets to tackle complex problems and use approximate methods like Approximate Bayesian Computation (ABC) or Bayesian Synthetic Likelihood (BSL) to accelerate computation. However, the number of simulations required remains a limiting factor.
Article
Genetics & Heredity
Alessandra Modi, Maria Teresa Vizzari, Giulio Catalano, Rajiv Boscolo Agostini, Stefania Vai, Martina Lari, Chiara Vergata, Valentina Zaro, Lucia Liccioli, Mariaelena Fedi, Serena Barone, Lorenzo Nigro, Hovirag Lancioni, Alessandro Achilli, Luca Sineo, David Caramelli, Silvia Ghirotto
Summary: Sicily, as one of the main islands of the Mediterranean Sea, has a rich history of migrations and populations' interaction, shaping its genetic landscape and cultural dynamics. The analysis of mitochondrial genomes from individuals spanning from the Early Bronze Age to the Iron Age in different locations on the island revealed structured genetic variation since the Early Bronze Age, pointing towards a demic impact of cultural transitions within Sicily. The modern genetic data of Sicilian mitochondrial DNA suggest a recent genetic replacement of pre-Iron Age populations, warranting further investigation.
FRONTIERS IN GENETICS
(2022)
Article
Physics, Multidisciplinary
Mijung Park, Margarita Vinaroz, Wittawat Jitkrittum
Summary: The ABCDP framework combines sparse vector technique to produce differentially private and approximate posterior samples in ABC, reducing cumulative privacy loss and providing higher privacy levels.
Article
Infectious Diseases
Sara Raimondi, Sara Gandini, Gibran Horemheb Rubio Quintanares, Ana Abecasis, Pier Luigi Lopalco, Oriana D'Ecclesiis, Susanna Chiocca, Elisa Tomezzoli, Ilaria Cutica, Davide Mazzoni, Nuno Amparo, Marta Pingarilho, Daniela Carmagnola, Claudia Dallavia, Gianvincenzo Zuccotti, Chiara Ronchini, Federica Bellerba, Felix Dewald, Rolf Kaiser, Francesca Incardona
Summary: This study aims to determine whether the Lolli-Methode (LM) is useful in supporting schools reopening and reducing clusters and attack rates of SARS-CoV-2. The study will enroll 440 classes from two countries, with samples collected and tested using PCR techniques. An observational study will also be conducted to evaluate the effectiveness of preventive measures and psychological issues in students and teachers.
BMC INFECTIOUS DISEASES
(2023)
Article
Biochemistry & Molecular Biology
David Pires, Manoj Mandal, Jacinta Pinho, Maria Joao Catalao, Antonio Jose Almeida, Jose Miguel Azevedo-Pereira, Maria Manuela Gaspar, Elsa Anes
Summary: Mycobacterium tuberculosis establishes chronic colonization in lung macrophages through controlled replication, leading to latent infection. Manipulation of endolysosomal enzymes, cathepsins, by the pathogen is crucial for intracellular survival. Liposomal delivery of saquinavir, a protease inhibitor, enhances its internalization in macrophages and exhibits significant intracellular killing effects on drug-susceptible and drug-resistant Mycobacterium tuberculosis strains.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Virology
Martin Pirkl, Joachim Buech, Carole Devaux, Michael Boehm, Anders Sonnerborg, Francesca Incardona, Ana Abecasis, Anne-Mieke Vandamme, Maurizio Zazzi, Rolf Kaiser, Thomas Lengauer
Summary: Human immunodeficiency virus (HIV) can develop resistance to all antiretroviral drugs, but multidrug resistance is rare. The accumulation of resistance mutations follows a predetermined order rather than being acquired randomly and simultaneously. Understanding this order can help elucidate the mechanisms of multidrug resistance.
JOURNAL OF MEDICAL VIROLOGY
(2023)
Article
Pharmacology & Pharmacy
Filipa Santos, David Pires, Elsa Anes, Ana Rita C. Duarte
Summary: This study evaluated the stability and antibacterial activity of therapeutic liquid formulations prepared with anti-tuberculosis drugs. The results showed that these mixtures have antibacterial effects against drug-susceptible Mycobacterium tuberculosis strains, with the mixtures incorporating ethambutol showing particularly prominent effects. The findings suggest the potential for further research and evaluation of clinical applicability.
INTERNATIONAL JOURNAL OF PHARMACEUTICS
(2023)
Article
Virology
Marta Calado, David Pires, Carolina Conceicao, Rita Ferreira, Quirina Santos-Costa, Elsa Anes, Jose Miguel Azevedo-Pereira
Summary: Macrophages and dendritic cells are important for the spread of HIV to CD4+ T lymphocytes during acute infection and constitute a persistently infected reservoir during chronic infection. Cell-to-cell contact triggers the production of infectious viral particles, contributing to viral replication. The phenotypic characteristics of HIV isolates do not correlate with their spread or the difference between HIV-1 and HIV-2 in terms of cis- or trans-infection. Understanding the cell-to-cell spread of HIV is critical for developing new therapeutic and vaccine approaches.
Article
Immunology
Catia Silveiro, Mariana Marques, Francisco Olivenca, David Pires, Diana Mortinho, Alexandra Nunes, Madalena Pimentel, Elsa Anes, Maria Joao Catalao
Summary: The study investigates the impact of the lack of effective therapeutics on multi-drug resistant strains of Mycobacterium tuberculosis. It identifies the essentiality of peptidoglycan modifications and their effects on resistance and host-pathogen interactions. Depletion of these modifications enhances the killing of bacteria by macrophages and shows potential as therapeutic targets against TB.
FRONTIERS IN CELLULAR AND INFECTION MICROBIOLOGY
(2023)
Review
Biochemistry & Molecular Biology
Elsa Anes, David Pires, Manoj Mandal, Jose Miguel Azevedo-Pereira
Summary: Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis, has developed a unique strategy to establish infection and transmission by utilizing host macrophages to establish intracellular niches and induce long-term latency in granulomas. It eventually escapes from macrophages through necrotic cell death and triggers a strong inflammatory response, which is necessary for the progression from latency to active disease and transmission.
Article
Microbiology
Joao P. Pais, Olha Antoniuk, Raquel Freire, David Pires, Emilia Valente, Elsa Anes, Luis Constantino
Summary: In this study, a library of 72 derivatives of nitrobenzoic acid containing esters and thioesters of benzoates was synthesized and evaluated for their antimycobacterial activity against M. tuberculosis. The results showed that compounds with aromatic nitro substitutions, especially 3,5-dinitro esters, exhibited the most potent antimicrobial activity. Interestingly, the activity of the nitro derivatives was not correlated with their pKa values or hydrolysis rates. These findings suggest that the nitrobenzoate scaffold, particularly the 3,5-dinitrobenzoate scaffold, has the potential to serve as a promising scaffold for the development of novel antimycobacterial agents with improved activity.
Review
Microbiology
Jose Miguel Azevedo-Pereira, David Pires, Marta Calado, Manoj Mandal, Quirina Santos-Costa, Elsa Anes
Summary: Human immunodeficiency virus (HIV) and Mycobacterium tuberculosis (Mtb) are responsible for millions of new infections annually, causing high morbidity and mortality globally. HIV increases the risk of tuberculosis (TB) development by a factor of 20, even in latently infected individuals, and controlled HIV infection on antiretroviral therapy (ART) still results in a fourfold increased risk of TB. Conversely, Mtb infection worsens HIV pathogenesis and accelerates AIDS progression. Understanding the reciprocal amplification of HIV/Mtb coinfection can aid in the development of therapeutic strategies for disease control, particularly in situations where vaccines or pathogen clearance are not readily available.
Review
Virology
Marta Calado, David Pires, Carolina Conceicao, Quirina Santos-Costa, Elsa Anes, Jose Miguel Azevedo-Pereira
Summary: Despite the success of antiretroviral therapy, the global HIV infection rate remains high, mainly through sexual transmission involving specific cells present on genital mucosa. Understanding how HIV interacts with these cells is crucial for developing prevention and control strategies. This review explores the manipulation of physiological roles of dendritic cells, macrophages, and CD4+ T lymphocytes by HIV in order to establish infection in a new human host. The interactions between HIV and these cells, including intercellular viral transfer mechanisms, are discussed.
REVIEWS IN MEDICAL VIROLOGY
(2023)
Article
Virology
Leonid Joaquim, Mafalda N. S. Miranda, Victor Pimentel, Maria do Rosario Oliveira Martins, Tacilta Nhampossa, Ana Abecasis, Marta Pingarilho
Summary: This study analyzed the retention in clinical care and virological response of HIV-1 patients receiving ART at the Maputo Military Hospital (MMH). The results showed that age, unemployment, CD4 count, and hemoglobin levels were associated with patient loss to follow-up and virological failure.
Article
Virology
Franziska Hufsky, Ana B. Abecasis, Artem Babaian, Sebastian Beck, Liam Brierley, Simon Dellicour, Christian Eggeling, Santiago F. Elena, Udo Gieraths, Anh D. Ha, Will Harvey, Terry C. Jones, Kevin Lamkiewicz, Gabriel L. Lovate, Dominik Luecking, Martin Machyna, Luca Nishimura, Maximilian K. Nocke, Bernard Y. Renard, Shoichi Sakaguchi, Lygeri Sakellaridi, Jannes Spangenberg, Maria Tarradas-Alemany, Sandra Triebel, Yulia Vakulenko, Rajitha Yasas Wijesekara, Fernando Gonzalez-Candelas, Sarah Krautwurst, Alba Perez-Cataluna, Walter Randazzo, Gloria Sanchez, Manja Marz
Summary: The 2023 International Virus Bioinformatics Meeting held in Valencia, Spain was a significant event for researchers and scientists worldwide interested in virus bioinformatics. With the primary objective of fostering discussions and collaborations, it provided a platform for sharing insights, research findings, and novel ideas related to virus bioinformatics research.
