scholarly journals Temporal state change Bayesian networks for modeling of evolving multivariate state sequences: model, structure discovery and parameter estimation

2021 ◽  
Author(s):  
Artur Mrowca ◽  
Florian Gyrock ◽  
Stephan Günnemann
Keyword(s):  
Parameter Estimation ◽  
Bayesian Networks ◽  
Probabilistic Models ◽  
Temporal Dynamics ◽  
Likelihood Estimation ◽  
State Change ◽  
Specification Mining ◽  
Partially Observed ◽  
Discovery Method ◽  
Over Time

AbstractMany systems can be expressed as multivariate state sequences (MSS) in terms of entities and their states with evolving dependencies over time. In order to interpret the temporal dynamics in such data, it is essential to capture relationships between entities and their changes in state and dependence over time under uncertainty. Existing probabilistic models do not explicitly model the evolution of causality between dependent state sequences and mostly result in complex structures when representing complete causal dependencies between random variables. To solve this, Temporal State Change Bayesian Networks (TSCBN) are introduced to effectively model interval relations of MSSs under evolving uncertainty. Our model outperforms competing approaches in terms of parameter complexity and expressiveness. Further, an efficient structure discovery method for TSCBNs is presented, that improves classical approaches by exploiting temporal knowledge and multiple parameter estimation approaches for TSCBNs are introduced. Those are expectation maximization, variational inference and a sampling based maximum likelihood estimation that allow to learn parameters from partially observed MSSs. Lastly, we demonstrate how TSCBNs allow to interpret and infer patterns of captured sequences for specification mining in automotive.

scholarly journals Acute symptoms of mild to moderate COVID-19 are highly heterogeneous across individuals and over time

2021 ◽  
Author(s):  
Thomas L Rodebaugh ◽  
Madelyn R Frumkin ◽  
Angela M Reiersen ◽  
Eric J Lenze ◽  
Michael S Avidan ◽  
...  
Keyword(s):  
Temporal Dynamics ◽  
Survey Methods ◽  
Clinical Care ◽  
Clinical Deterioration ◽  
Symptom Profiles ◽  
Fluctuating Course ◽  
The Common ◽  
Symptom Changes ◽  
Over Time

Abstract Background The symptoms of COVID-19 appear to be heterogenous, and the typical course of these symptoms is unknown. Our objectives were to characterize the common trajectories of COVID-19 symptoms and assess how symptom course predicts other symptom changes as well as clinical deterioration. Methods 162 participants with acute COVID-19 responded to surveys up to 31 times for up to 17 days. Several statistical methods were used to characterize the temporal dynamics of these symptoms. Because nine participants showed clinical deterioration, we explored whether these participants showed any differences in symptom profiles. Results Trajectories varied greatly between individuals, with many having persistently severe symptoms or developing new symptoms several days after being diagnosed. A typical trajectory was for a symptom to improve at a decremental rate, with most symptoms still persisting to some degree at the end of the reporting period. The pattern of symptoms over time suggested a fluctuating course for many patients. Participants who showed clinical deterioration were more likely to present with higher reports of severity of cough and diarrhea. Conclusion The course of symptoms during the initial weeks of COVID-19 is highly heterogeneous and is neither predictable nor easily characterized using typical survey methods. This has implications for clinical care and early-treatment clinical trials. Additional research is needed to determine whether the decelerating improvement pattern seen in our data is related to the phenomenon of patients reporting long-term symptoms, and whether higher symptoms of diarrhea in early illness presages deterioration.

scholarly journals Longitudinal Patterns of Ethical Organisational Culture as a Context for Leaders’ Well-Being: Cumulative Effects Over 6 Years

2021 ◽  
Author(s):  
Mari Huhtala ◽  
Muel Kaptein ◽  
Joona Muotka ◽  
Taru Feldt
Keyword(s):  
Latent Profile Analysis ◽  
Temporal Dynamics ◽  
Ethical Dilemmas ◽  
Extended Period ◽  
Well Being ◽  
Organisational Culture ◽  
Cumulative Effects ◽  
Ethical Culture ◽  
Continuous Increase ◽  
Over Time

AbstractThe aim of this longitudinal study was to investigate the temporal dynamics of ethical organisational culture and how it associates with well-being at work when potential changes in ethical culture are measured over an extended period of 6 years. We used a person-centred study design, which allowed us to detect both typical and atypical patterns of ethical culture stability as well as change among a sample of leaders. Based on latent profile analysis and hierarchical linear modelling we found longitudinal, concurrent relations and cumulative gain and loss cycles between different ethical culture patterns and leaders’ well-being. Leaders in the strongest ethical culture pattern experienced the highest level of work engagement and a decreasing level of ethical dilemmas and stress. Leaders who gave the lowest ratings on ethical culture which also decreased over time reported the highest level of ethical dilemmas, stress, and burnout. They also showed a continuous increase in these negative outcomes over time. Thus, ethical culture has significant cumulative effects on well-being, and these longitudinal effects can be both negative and positive, depending on the experienced strength of the culture’s ethicality.

scholarly journals 520. Longitudinal Analysis of SARS-CoV-2 Viruses in Hospitalized Adults

2020 ◽  
Vol 7 (Supplement_1) ◽  
pp. S325-S326
Author(s):  
Lacy Simons ◽  
Ramon Lorenzo-Redondo ◽  
Hannah Nam ◽  
Scott C Roberts ◽  
Michael G Ison ◽  
...  
Keyword(s):  
Viral Load ◽  
Genome Sequencing ◽  
Viral Genome ◽  
Temporal Dynamics ◽  
Population Level ◽  
Hospital Staff ◽  
Therapeutic Interventions ◽  
Viral Loads ◽  
Qrt Pcr ◽  
Over Time

Abstract Background The rapid spread of SARS-CoV-2, the causative agent of Coronavirus disease 2019 (COVID-19), has been accompanied by the emergence of viral mutations, some of which may have distinct virological and clinical consequences. While whole genome sequencing efforts have worked to map this viral diversity at the population level, little is known about how SARS-CoV-2 may diversify within a host over time. This is particularly important for understanding the emergence of viral resistance to therapeutic interventions and immune pressure. The goal of this study was to assess the change in viral load and viral genome sequence within patients over time and determine if these changes correlate with clinical and/or demographic parameters. Methods Hospitalized patients admitted to Northwestern Memorial Hospital with a positive SARS-CoV-2 test were enrolled in a longitudinal study for the serial collection of nasopharyngeal specimens. Swabs were administered to patients by hospital staff every 4 ± 1 days for up to 32 days or until the patients were discharged. RNA was extracted from each specimen and viral loads were calculated by quantitative reverse transcriptase PCR (qRT-PCR). Specimens with qRT-PCR cycle threshold values less than or equal to 30 were subject to whole viral genome sequencing by reverse transcription, multiplex PCR, and deep sequencing. Variant populations sizes were estimated and subject to phylogenetic analysis relative to publicly available SARS-CoV-2 sequences. Sequence and viral load data were subsequently correlated to available demographic and clinical data. Results 60 patients were enrolled from March 26th to June 20th, 2020. We observed an overall decrease in nasopharyngeal viral load over time across all patients. However, the temporal dynamics of viral load differed on a patient-by-patient basis. Several mutations were also observed to have emerged within patients over time. Distribution of SARS-CoV-2 viral loads in serially collected nasopharyngeal swabs in hospitalized adults as determined by qRT-PCR. Samples were collected every 4 ± 1 days (T#1–8) and viral load is displayed by log(copy number). Conclusion These data indicate that SARS-CoV-2 viral loads in the nasopharynx decrease over time and that the virus can accumulate mutations during replication within individual patients. Future studies will examine if some of these mutations may provide fitness advantages in the presence of therapeutic and/or immune selective pressures. Disclosures Michael G. Ison, MD MS, AlloVir (Consultant)

l2-Penalized temporal logit-mixed models for the estimation of regional obesity prevalence over time

2021 ◽  
pp. 096228022110175
Author(s):  
Jan P Burgard ◽  
Joscha Krause ◽  
Ralf Münnich ◽  
Domingo Morales
Keyword(s):  
Parameter Estimation ◽  
Medical Treatment ◽  
Mixed Models ◽  
Generalized Linear Mixed Models ◽  
Linear Mixed Models ◽  
Obesity Prevalence ◽  
Model Parameter ◽  
Model Parameter Estimation ◽  
Public Health Reporting ◽  
Over Time

Obesity is considered to be one of the primary health risks in modern industrialized societies. Estimating the evolution of its prevalence over time is an essential element of public health reporting. This requires the application of suitable statistical methods on epidemiologic data with substantial local detail. Generalized linear-mixed models with medical treatment records as covariates mark a powerful combination for this purpose. However, the task is methodologically challenging. Disease frequencies are subject to both regional and temporal heterogeneity. Medical treatment records often show strong internal correlation due to diagnosis-related grouping. This frequently causes excessive variance in model parameter estimation due to rank-deficiency problems. Further, generalized linear-mixed models are often estimated via approximate inference methods as their likelihood functions do not have closed forms. These problems combined lead to unacceptable uncertainty in prevalence estimates over time. We propose an l2-penalized temporal logit-mixed model to solve these issues. We derive empirical best predictors and present a parametric bootstrap to estimate their mean-squared errors. A novel penalized maximum approximate likelihood algorithm for model parameter estimation is stated. With this new methodology, the regional obesity prevalence in Germany from 2009 to 2012 is estimated. We find that the national prevalence ranges between 15 and 16%, with significant regional clustering in eastern Germany.

scholarly journals Markov-Modulated Continuous-Time Markov Chains to Identify Site- and Branch-Specific Evolutionary Variation in BEAST

2020 ◽  
Vol 70 (1) ◽  
pp. 181-189
Author(s):  
Guy Baele ◽  
Mandev S Gill ◽  
Paul Bastide ◽  
Philippe Lemey ◽  
Marc A Suchard
Keyword(s):  
High Performance ◽  
Markov Models ◽  
Marginal Likelihood ◽  
Likelihood Estimation ◽  
Phylogenetic Inference ◽  
Time Variability ◽  
Substitution Process ◽  
Wide Range ◽  
Markov Modulated ◽  
Over Time

Abstract Markov models of character substitution on phylogenies form the foundation of phylogenetic inference frameworks. Early models made the simplifying assumption that the substitution process is homogeneous over time and across sites in the molecular sequence alignment. While standard practice adopts extensions that accommodate heterogeneity of substitution rates across sites, heterogeneity in the process over time in a site-specific manner remains frequently overlooked. This is problematic, as evolutionary processes that act at the molecular level are highly variable, subjecting different sites to different selective constraints over time, impacting their substitution behavior. We propose incorporating time variability through Markov-modulated models (MMMs), which extend covarion-like models and allow the substitution process (including relative character exchange rates as well as the overall substitution rate) at individual sites to vary across lineages. We implement a general MMM framework in BEAST, a popular Bayesian phylogenetic inference software package, allowing researchers to compose a wide range of MMMs through flexible XML specification. Using examples from bacterial, viral, and plastid genome evolution, we show that MMMs impact phylogenetic tree estimation and can substantially improve model fit compared to standard substitution models. Through simulations, we show that marginal likelihood estimation accurately identifies the generative model and does not systematically prefer the more parameter-rich MMMs. To mitigate the increased computational demands associated with MMMs, our implementation exploits recent developments in BEAGLE, a high-performance computational library for phylogenetic inference. [Bayesian inference; BEAGLE; BEAST; covarion, heterotachy; Markov-modulated models; phylogenetics.]

Classification Based on Multivariate Repeated Measures with Time Effect on Mean Vector and an AR (1) Correlation Structure on the Repeated Measures

2005 ◽  
Vol 57 (1-2) ◽  
pp. 49-66 ◽  
Author(s):  
Anuradba Roy ◽  
Ravindra Khattree
Keyword(s):  
Maximum Likelihood ◽  
Repeated Measures ◽  
Likelihood Estimation ◽  
Covariance Matrices ◽  
Time Effect ◽  
Population Parameters ◽  
Time Effects ◽  
Mean Vectors ◽  
Mean Vector ◽  
Over Time

In repeated measures studies how observations change over time is often of prime interest. Modelling this time effect in the context of discrimination, is the objective of this article. We study the problem of classification with multiple q-variate observations with time effect on each individual. The covariance matrices as well as mean vectors are mordelled respectively to accommodate the correlation between the successive repeated measures and to describe the time effects. Computation schemes for maximum likelihood estimation of required population parameters are provided.

scholarly journals Temporal Dynamics of Cloacal Microbiota in Adult Laying Chickens With and Without Access to an Outdoor Range

2021 ◽  
Vol 11 ◽  
Author(s):  
Janneke Schreuder ◽  
Francisca C. Velkers ◽  
Alex Bossers ◽  
Ruth J. Bouwstra ◽  
Willem F. de Boer ◽  
...  
Keyword(s):  
Microbial Community ◽  
Community Composition ◽  
Intestinal Microbiota ◽  
Environmental Changes ◽  
Temporal Dynamics ◽  
Microbial Community Composition ◽  
Sudden Change ◽  
Rrna Gene ◽  
Poultry House ◽  
Over Time

Associations between animal health and performance, and the host’s microbiota have been recently established. In poultry, changes in the intestinal microbiota have been linked to housing conditions and host development, but how the intestinal microbiota respond to environmental changes under farm conditions is less well understood. To gain insight into the microbial responses following a change in the host’s immediate environment, we monitored four indoor flocks of adult laying chickens three times over 16 weeks, during which two flocks were given access to an outdoor range, and two were kept indoors. To assess changes in the chickens’ microbiota over time, we collected cloacal swabs of 10 hens per flock and performed 16S rRNA gene amplicon sequencing. The poultry house (i.e., the stable in which flocks were housed) and sampling time explained 9.2 and 4.4% of the variation in the microbial community composition of the flocks, respectively. Remarkably, access to an outdoor range had no detectable effect on microbial community composition, the variability of microbiota among chickens of the same flock, or microbiota richness, but the microbiota of outdoor flocks became more even over time. Fluctuations in the composition of the microbiota over time within each poultry house were mainly driven by turnover in rare, rather than dominant, taxa and were unique for each flock. We identified 16 amplicon sequence variants that were differentially abundant over time between indoor and outdoor housed chickens, however none were consistently higher or lower across all chickens of one housing type over time. Our study shows that cloacal microbiota community composition in adult layers is stable following a sudden change in environment, and that temporal fluctuations are unique to each flock. By exploring microbiota of adult poultry flocks within commercial settings, our study sheds light on how the chickens’ immediate environment affects the microbiota composition.

scholarly journals Stochastic Modeling Reveals Kinetic Heterogeneity in Post-replication DNA Methylation

2019 ◽  
Author(s):  
Luis Busto-Moner ◽  
Julien Morival ◽  
Arjang Fahim ◽  
Zachary Reitz ◽  
Timothy L. Downing ◽  
...  
Keyword(s):  
Mathematical Modeling ◽  
Dna Methylation ◽  
Rate Constants ◽  
Cytosine Methylation ◽  
Temporal Dynamics ◽  
Epigenetic Modification ◽  
Embryonic Stem ◽  
Likelihood Estimation ◽  
Dna Methyltransferases ◽  
Methylation Patterns

AbstractDNA methylation is a heritable epigenetic modification that plays an essential role in mammalian development. Genomic methylation patterns are dynamically maintained, with DNA methyltransferases mediating inheritance of methyl marks onto nascent DNA over cycles of replication. A recently developed experimental technique employing immunoprecipitation of bromodeoxyuridine labeled nascent DNA followed by bisulfite sequencing (Repli-BS) measures post-replication temporal evolution of cytosine methylation, thus enabling genome-wide monitoring of methylation maintenance. In this work, we combine statistical analysis and stochastic mathematical modeling to analyze Repli-BS data from human embryonic stem cells. We estimate site-specific kinetic rate constants for the restoration of methyl marks on >10 million uniquely mapped cytosines within the CpG (cytosine-phosphate-guanine) dinucleotide context across the genome using Maximum Likelihood Estimation. We find that post-replication remethylation rate constants span approximately two orders of magnitude, with half-lives of per-site recovery of steady-state methylation levels ranging from shorter than ten minutes to five hours and longer. Furthermore, we find that kinetic constants of maintenance methylation are correlated among neighboring CpG sites. Stochastic mathematical modeling provides insight to the biological mechanisms underlying the inference results, suggesting that enzyme processivity and/or collaboration can produce the observed kinetic correlations. Our combined statistical/mathematical modeling approach expands the utility of genomic datasets and disentangles heterogeneity in methylation patterns arising from replication-associated temporal dynamics versus stable cell-to-cell differences.

Sign in / Sign up

Export Citation Format

Share Document