Bayesian Inference Research Articles

A Review on Bridging CAM and Modern Statistical Tools: An Introduction to Bayesian Methods

Complementary and Alternative Medicine (CAM) modalities, encompassing practices such as acupuncture, yoga, and herbal therapy, are experiencing a notable increase in their utilization as supplementary or alternative options to conventional medical interventions. Nonetheless, the scientific appraisal of these modalities is frequently hindered by methodological issues, including the personalization of treatments, limited sample sizes, and multifaceted, multidimensional outcomes. Such constraints often render traditional frequentist statistical techniques inadequate for evaluating the efficacy of CAM interventions. This review highlights the potential of Bayesian statistical approaches as a transformative solution for the study of CAM.Through an extensive review of existing literature and the development of hypothetical models, we illustrate the advantages of Bayesian methodologies in tackling the specific challenges associated with CAM research. Review findings include the capacity to integrate prior knowledge derived from historical and observational data, thereby effectively addressing issues related to small sample sizes and facilitating adaptive trial designs that enhance resource allocation.Furthermore, Bayesian hierarchical models adeptly accommodate heterogeneity across varied patient demographics and therapeutic modalities, while allowing for the seamless integration of multidimensional outcomes into a cohesive analytical framework. Despite challenges such as computational intensity and the subjective nature of prior selection, advancements in computational software and educational resources are improving the accessibility of Bayesian methodologies. This paradigm presents a robust, evidence-based framework for the validation of CAM therapies, thereby promoting their incorporation into clinical practice. By effectively addressing methodological intricacies, Bayesian statistics has the potential to elevate CAM research to align with the standards of contemporary evidence-based medicine, thereby encouraging wider acceptance within healthcare systems.

Online Bayesian State Estimation for Real-Time Monitoring of Growth Kinetics in Thin Film Synthesis.

Rapid validation of newly predicted materials through autonomous synthesis requires real-time adaptive control methods that exploit physics knowledge, a capability that is lacking in most systems. Here, we demonstrate an approach to enable real-time control of thin film synthesis by combining in situ optical diagnostics with a Bayesian state estimation method. We developed a physical model for film growth and applied the direct filter (DF) method for real-time estimation of nucleation and growth rates during pulsed laser deposition (PLD). We validated the approach using simulated and experimental reflectivity data for WSe2 growth and ultimately deployed the algorithm on an autonomous PLD system during the growth of 1T'-MoTe2. The DF robustly estimates growth parameters in real time at early stages of growth, down to 15% monolayer area coverage. This fusion of in situ diagnostics, data assimilation, and physical modeling opens new opportunities in adaptive control of synthesis trajectories toward desired material states.

Performance of weighted genomic BLUP and Bayesian methods for Hanwoo carcass traits.

To improve the quality and yield of the Korean beef industry, selection criteria often focus on estimated breeding values for carcass weight (CWT), eye muscle area (EMA), backfat thickness (BF), and marbling score (MS). This study estimated genetic parameters and assessed the accuracy of genomic estimated breeding values (GEBVs) using SNP weighting methods. We compared the accuracy of these methods with the genomic best linear unbiased prediction (GBLUP) and various Bayesian approaches (BayesA, BayesB, BayesC, and BayesCPi) for the specified traits. The study used single-trait animal models, including GBLUP, weighted GBLUP (WGBLUP), and the Bayesian methods to predict genomic breeding values in a population of Hanwoo steers. A total of 19154 phenotypes were collected with all animals genotyped using the Illumina Bovine 50K SNP chip. The average heritability for the carcass traits was 0.33 (GBLUP) and 0.35 (Bayesian), with Bayesian methods yielding heritability estimates that were on average 0.02 points (6.1%) higher than GBLUP. The accuracy of genomic predictions ranged from 0.7-0.83 (GBLUP), 0.83-0.87 (WGBLUP), and 0.81-0.87 across the Bayesian methods. WGBLUP accuracies for the carcass traits were, on average 8.97% higher than the GBLUP accuracies and 1.80% higher than the Bayesian alphabets. The Bayesian alphabet's accuracy is also, on average 6.00% higher than the GBLUP accuracy. According to these findings, the weighting GBLUP approach provides higher prediction accuracy for Hanwoo carcass traits than the Bayesian alphabet. Therefore, WGBLUP can be used for genomic selection in the Hanwoo evaluation program.

Driving style recognition based on a Bayesian belief-renewing method

Developing connected and automated vehicle (CAV) technologies that can interact effectively with surrounding intelligent agents is a crucial challenge to cope with the near-future mixed-flow environment composed of CAVs and human-driven vehicles (HDVs). A fundamental problem to be solved is understanding and recognising driving styles, which can help CAVs reason and predict human driving behaviour. Motivated by it, this study designs a belief-renewing method to recognise the styles of different driving primitives. We first propose a hierarchical Dirichlet process hidden semi-Markov model (HDP-VAR-GP-HSMM) considering time-varying properties to extract driving primitives from natural driving behaviours. The extracted primitives are divided into two driving styles (aggressive or cautious) to build a training dataset with a multi-dimensional k-means clustering algorithm. Next, a Long Short-Term Memory (LSTM) neural network with the Bayesian posterior correction model is trained to recognise the driving styles of primitives. The recognition model can dynamically renew the belief of CAVs on target vehicles' driving styles to mitigate the negative effect caused by stochastic fluctuations. The proposed algorithm is tested using an open-source freeway merging zone dataset. Experimental results show that: (1) the proposed HDP-VAR-GP-HSMM can capture time-varying characteristics of driving primitives and achieves better performance compared with other nonparametric approaches, (2) the accuracy of recognition models with Bayesian belief-renewing method is higher than conventional models. Related approaches and findings in this study can be helpful in the decision-making and control of CAVs.

BART-Survival: A Bayesian machine learning approach to survival analyses in Python

BART-Survival: A Bayesian machine learning approach to survival analyses in Python

Quantum Bayesian Inference with Renormalization for Gravitational Waves

Quantum Bayesian Inference with Renormalization for Gravitational Waves

Proton response and neutron spectrum unfolding by solution-grown trans-stilbene scintillator

This paper reports the proton response function for solution-grown trans-stilbene scintillator from 1 to 25 MeVee and its application for unfolding neutron spectra of Cf-252 and AmBe sources. Low energy proton response was measured by the Time-of-Flight technique at the Korea Research Institute of Standards and Science. In contrast, high energy response was investigated using a 45 MeV proton beam at the Korea Institute of Radiological & Medical Sciences. The photomultiplier tube nonlinearity effect was studied and corrected for high-energy measurement. Several energy calibration functions were examined to minimize the effects of the light yield nonlinearity of the scintillator, which was observed as nonnegligible in the measurement. A new approach was applied to obtain 1.3 ns full width at half maximum timing resolution in the Time-of-Flight experiment when using a 500 Mega samples per second analog to digital converter. The unfolding was performed using the iterative Bayesian method without any constraints, smoothness, or assumption of prior knowledge of original spectra. For the first time, we achieved one of the best unfolding results for both Cf-252 and AmBe spectra. Results from this study are promising for fast neutron spectroscopy.

Spatially explicit estimation of recent migration rates in plants using genotypic data.

We present a new hierarchical Bayesian method using multilocus genotypes to estimate recent seed and pollen migration rates in a spatially explicit framework that incorporates distance effects separately for each type of dispersal. The method additionally estimates population allelic frequencies, population divergence values, individual inbreeding coefficients, individual maternal and paternal ancestries, and allelic dropout rates. We conduct a numerical simulation analysis that indicates that the method can provide reliable estimates of seed and pollen migration rates and allow accurate inference of spatial effects on migration, at affordable sample sizes (25-50 individuals/population) when population genetic divergence is not low (FST≥0.05), or by increasing sampling (to at least 100 individuals/population) under weaker levels of divergence (FST=0.025). Simulations also show that the accuracy provided by assays with about one thousand unlinked polymorphic SNP loci may approach, for a given sample size, the theoretical maximum achievable under categorical origin discrimination. We apply our method to Taxus baccata data, revealing low but significant seed and pollen migration among nearby population remnants during the last generation, with a negative effect of interpopulation distance on migration that was detectable for pollen but not for seeds.

Information and communication technology use among social ties of varying strength: Who benefits and how?

Information and communication technology (ICT) use has been associated with well-being among older adults. This link is often attributed to the fact that ICT use facilitates connecting with others. The purpose of this study is to assess how contact frequency and social tie strength impact the relationship between ICT use and depressive symptoms among older adults. Using data from the Detroit-based Social Relations Study collected in 2015, we use Bayesian analyses to examine the extent to which contact frequency with social network members moderates and the strength of these social tie (strong, moderate, and weak) mediate the link between ICT use and depressive symptoms among adults age 60+ (n=483). ICT use was found to be associated with fewer depressive symptoms. Furthermore, ICT use was associated with more moderate and weak ties which were, in turn, associated with fewer depressive symptoms. Finally, a moderating effect was also found as ICT use was associated with fewer depressive symptoms only among those with lower contact frequency. Our findings suggest that for older adults who communicate with others less frequently, ICT use can prove beneficial to mental health (i.e. fewer depressive symptoms). Furthermore ICT use may be especially meaningful to connect with weaker social ties. These findings may be due in part to ICTs capabilities which enable older adults to maintain connection to a diverse array of social ties and bridge social and physical distance.

The longitudinal network of peer problems and emotional symptoms among Australian adolescents: Bayesian structure learning of directed acyclic graphs.

Adolescence is a period in which peer problems and emotional symptoms markedly increase in prevalence. However, the causal mechanisms regarding how peer problems cause emotional symptoms at a behavioral level and vice versa remain unknown. To address this gap, the present study investigated the longitudinal network of peer problems and emotional symptoms among Australian adolescents aged 12-14 years. Data were from the Longitudinal Study of Australian Children. The complete case samples included adolescents who participated in the Longitudinal Study of Australian Children B (n = 2,694) or K (n = 3,144) Cohorts at two study follow-ups (ages 12 and 14). Peer problems and emotional symptoms were measured with the self-report Strengths and Difficulties Questionnaire. The analytical steps were (a) in Study 1, a causal discovery algorithm, Bayesian structure learning of directed acyclic graphs (DAGs), was used to identify the longitudinal network in the K Cohort; (b) the DAG discovered was evaluated with Bayesian structural equation modeling in an independent sample (the B Cohort) and compared against a DAG established through expert knowledge; and (c) in Study 2, the longitudinal network was again evaluated but considered contemporaneous effects. The empirically discovered DAG provided a better explanation of independent data than the expert DAG. Based on the discovered DAG, several plausible causal effects were identified such as that being bullied at age 12 negatively affected popularity at age 14. This study provides new insights into potential causal effects established between peer problems and emotional symptoms among Australian adolescents aged 12-14 years. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Bayesian inference of wall torques for active Brownian particles

Abstract The motility of living things and synthetic self-propelled objects is often described using Active Brownian particles. To capture the interaction of these particles with their often complex environment, this model can be augmented with empirical forces or torques, for example, to describe their alignment with an obstacle or wall after a collision.
Here, we assess the quality of these empirical models by comparing their output predictions with trajectories of rod-shaped active particles that scatter sterically at a flat wall. We employ a classical least-squares method to evaluate the instantaneous torque. In addition, we lay out a Bayesian inference procedure to construct the posterior distribution of plausible model parameters. In contrast to the least squares fit, the Bayesian approach does not require orientational data of the active particle and can readily be applied to experimental tracking data.

Morphological and phylogenetic analyses reveal four novel species of Distoseptispora (Distoseptisporaceae, Distoseptisporales) from southern China

Saprobic hyphomycetes with high density and species diversity were observed on plant debris. During our mycological surveys in southern China, numerous strains were isolated from dead branches of unidentified plants in the forest of Jiangxi and Fujian provinces, China. Phylogenetic analyses of ITS, LSU, RPB2, and TEF1 sequence data using maximum-likelihood (ML) and Bayesian inference (BI) methods revealed the systematic placement of several Sporidesmium-like species within Distoseptispora. Based on morphological characteristics and molecular evidence, four new species of Distoseptispora, namely D. fujianensis, D. ganzhouensis, D. nanpingensis and D. subtropica, are introduced, and two known species, D. clematidis and D. yunjushanensis, are reported. This study enhances our knowledge of the species diversity of Distoseptispora in southern China.

The Beneficial Effects of Alpha-Blockers, Antimuscarinics, Beta 3-Agonist, and PDE5-Inhibitors for Ureteral Stent-Related Discomfort: A Systematic Review and Meta-Analysis from KSER Update Series

Background and Objectives: Ureteral stents are widely used in the field of urology but can cause varying degrees of side effects. This study utilized a network meta-analysis to evaluate stent-related discomfort (SRD) in patients with alpha-blockers (alfuzosin, tamsulosin, and silodosin), antimuscarinics (solifenacin), beta 3-agonists (mirabegron), and phosphodiesterase 5-inhibitors (tadalafil) versus a placebo. Materials and Methods: Relevant randomized controlled trials (RCTs) from 2006 to 2021 were identified from electronic databases, including PubMed, EMBASE, and the Cochrane Library. The following identifiers were included to assess the urinary symptom score (USS): participants (patients with ureteral stents), interventions (patients who took medication for stent discomfort), and outcomes (comparisons of the Ureteric Stent Symptoms Questionnaire (USSQ)). We also executed an independent quality assessment using the Scottish Intercollegiate Guidelines Network (SIGN). Results: A total of 16 RCTs were identified, and they included 1865 patients. Compared with the placebo, mirabegron (mean difference (MD): −3.87; 95% confidence interval (CI): −10.6–2.35), tadalafil (MD: −4.47; 95% CI: −10.8–1.63), and silodosin (MD: −4.02; 95% CI: −12–4.01) did not show significant differences to the placebo, whereas others did. Alfuzosin, mirabegron, silodosin, solifenacin, and tadalafil were not inferior to tamsulosin in terms of the USS using Bayesian analyses. In the random effect model, P-score tests showed that solifenacin possessed the highest P-score (p = 0.8484); tamsulosin was the second highest (p = 0.7054). As a result of the rank-probability test, solifenacin was also ranked highest in terms of USS, and tamsulosin was ranked second. Conclusions: Compared with the placebo, solifenacin, tamsulosin, and alfuzosin significantly decreased the USS. In our study, solifenacin may be considered the most effective medication for SRD.

Exploring Data Set Bias and Decision Support with Predictive Uncertainty Through Bayesian Approximations and Convolutional Neural Networks

Abstract Individual seismic catalogs can contain multiscale observations from fault level to global scales and associated waveforms from discrete events reflect crustal structure across many different scales and locations. Seismic network aperture, geographic location, and observation distance may not provide informative guidance or intuition on how different catalogs will behave across models trained under different conditions. We rely on uncertainty to provide guardrails for when to trust model decisions, but understanding when our uncertainty is trustworthy is an open challenge. Here, we explore Bayesian approximation methods for assigning predictive uncertainty in seismic event classification problems. We find that computationally expensive Bayesian approximations do not outperform simple ensemble methods. We also find that when exploiting multiple seismic event catalogs, joint training with data from all the catalogs combined with Bayesian approximations and supervised training for classification can obscure bias and result in less robust uncertainty while also not providing substantial performance benefits compared to training individual models for each catalog.

HIV Molecular Transmission Networks Among Students in Guangxi: Unraveling the Dynamics of Student-Driven HIV Epidemic

: In Guangxi, the number of newly diagnosed HIV-1 infections among students is continuously increasing, highlighting the need for a detailed understanding of local transmission dynamics, particularly focusing on key drivers of transmission. We recruited individuals newly diagnosed with HIV-1 in Nanning, Guangxi, and amplified and sequenced the HIV-1 pol gene to construct a molecular network. Bayesian phylogenetic analysis was utilized to identify migration events, and multivariable logistic regression was employed to analyze factors influencing clustering and high linkage. The predominant subtype among students was CRF07_BC (58.5%), followed by CRF01_AE (17.4%) and CRF55_01B (13.5%). Transmission network analysis identified a significant clustering rate of 64.3% among students, primarily within large clusters. The strongest transmission relationships were observed between students and MSM aged 25 to 39, as well as nonstudent youths. These migration events primarily occurred from MSM aged 25-39 to students and nonstudent youths for CRF01_AE, CRF07_BC, and CRF55_01B. Qingxiu was the main emigration region for for CRF01_AE, CRF07_BC, while Xixiangtang for CRF55_01B. Link with nonstudent youths (AOR = 5.11) and MSM aged 25 to 39 (AOR = 8.82) were significant factors contributing to the high linkage among students. Long-term infection was a key factor in super spreaders. These findings emphasize the critical role of MSM aged 25-39 in HIV-1 transmission among local youths, particularly regarding long-term infected individuals. The study advocates for targeted HIV-1 screening and intervention strategies for youths to strengthen early detection and treatment, thereby mitigating further transmission within this high-risk group.

Penalty kicks: an adversarial risk analysis (ARA) perspective

Abstract Penalty kicks are critical to game outcomes in soccer. The typical quantitative strategy is as follows. First, model such an interaction as a two or three strategy game between the kicker and the goalkeeper. Second, find the mixed-strategy Nash equilibrium (MSNE) to determine the “optimal” probabilities of each player choosing to kick or dive to either side or to the center of the goal. While this is the usual path to a solution, it is also fraught with many assumptions due to the nature of penalty kick data as well as due to implicit assumptions within the game theory model. In this paper, we introduce an alternative set of strategies, known as adversarial risk analysis (ARA), to determine optimal decisions for the penalty kick game. ARA, which is grounded in the principles of Bayesian decision analysis, allows the decision maker to avoid many of the assumptions and limitations encountered when using the game theory approach. By examining 2018–2019 Major League Soccer (MLS) penalty kicks on an individual basis, we show that our most accurate ARA model predicts the correct goalkeeper decision 65 % of the time, while the game-theoretic model predicts correctly 51 % of the time, a statistically significant difference.

Multiomic analyses direct hypotheses for Creutzfeldt-Jakob disease risk genes.

Prions are assemblies of misfolded prion protein that cause several fatal and transmissible neurodegenerative diseases, with the most common phenotype in humans being sporadic Creutzfeldt-Jakob disease (sCJD). Aside from variation of the prion protein itself, molecular risk factors are not well understood. Prion and prion-like mechanisms are thought to underpin common neurodegenerative disorders meaning that the elucidation of mechanisms could have broad relevance. Herein we sought to further develop our understanding of the factors that confer risk of sCJD using a systematic gene prioritization and functional interpretation pipeline based on multiomic integrative analyses. We integrated the published sCJD genome-wide association study (GWAS) summary statistics with publicly available bulk brain and brain cell type gene and protein expression datasets. We performed multiple transcriptome and proteome-wide association studies (TWAS & PWAS) and Bayesian genetic colocalization analyses between sCJD risk association signals and multiple brain molecular quantitative trait loci signals. We then applied our systematic gene prioritization pipeline on the obtained results and nominated prioritized sCJD risk genes with risk-associated molecular mechanisms in a transcriptome and proteome-wide manner. Genetic upregulation of both gene and protein expression of syntaxin-6 (STX6) in the brain was associated with sCJD risk in multiple datasets, with a risk-associated gene expression regulation specific to oligodendrocytes. Similarly, increased gene and protein expression of protein disulfide isomerase family A member 4 (PDIA4), involved in the unfolded protein response, was linked to increased disease risk, particularly in excitatory neurons. Protein expression of mesencephalic astrocyte derived neurotrophic factor (MANF), involved in protection against endoplasmic reticulum stress and sulfatide binding (linking to the enzyme in the final step of sulfatide synthesis, encoded by sCJD risk gene GAL3ST1), was identified as protective against sCJD. In total 32 genes were prioritized into two tiers based on the level of evidence and confidence for further studies. This study provides insights into the genetically-associated molecular mechanisms underlying sCJD susceptibility and prioritizes several specific hypotheses for exploration beyond the prion protein itself and beyond the previously highlighted sCJD risk loci through the newly prioritized sCJD risk genes and mechanisms. These findings highlight the importance of glial cells, sulfatides and the excitatory neuron unfolded protein response in sCJD pathogenesis.

BayesBay: A Versatile Bayesian Inversion Framework Written in Python

Abstract The lack of versatile tools for Bayesian inference presents a significant challenge to researchers in geophysics, who often resort to developing bespoke codes to address specific classes of inverse problems. In this study, we present BayesBay, a Python package for generalized transdimensional and hierarchical Markov chain Monte Carlo sampling. Leveraging object-oriented programming principles, BayesBay facilitates the definition of Bayesian sampling problems across a range of applications. This includes joint inversions of multiple data sets with different forward functions and unknown noise properties, as well as complex parameterizations involving multiple parameters with unknown dimensionality and/or spatially varying priors. We illustrate BayesBay from both a technical and a practical perspective. The first two applications are common in geophysics: a 2D tomographic problem and a joint inversion for the 1D subsurface structure. The third involves partition modeling and requires a sophisticated parameterization with two nested levels of transdimensionality. In all cases, BayesBay recovers known solutions, highlighting its potential to address a broad range of inverse problems.

Bayesian method for comparing F1 scores in the absence of a gold standard

ABSTRACT In the field of medicine, evaluating the diagnostic performance of new diagnostic methods can be challenging, especially in the absence of a gold standard. This study proposes a methodology for assessing the performance of diagnostic tests by estimating the posterior distribution of the F 1 score using latent class analysis, without relying on a gold standard. The proposed method utilizes Markov Chain Monte Carlo sampling to estimate the posterior distribution of the F 1 score, enabling a comprehensive evaluation of diagnostic test methods. By applying this method to internet addiction, we demonstrate how latent class analysis can be effectively used to assess diagnostic performance, offering a practical solution for situations where no gold standard is available. The effectiveness of the proposed approach was evaluated through simulation studies by examining the coverage probability of the 95% highest density interval of the estimated posterior distributions.

Cognitive ecology of surprise in predator–prey interactions

Abstract In this review, we relate theoretical work on the importance of surprise in cognition to empirical research relevant to surprise in predator–prey interactions. There have been multiple proposals as to how surprise should be defined and quantified in the context of animal cognition, including contributions from associative learning, information theory, Bayesian inference and the recent framework of active inference. We argue that active inference provides a novel and powerful approach to quantifying surprise and advances the field by revealing how proactive behaviour on the part of predators relates to reducing surprise. The active inference framework encompasses both proximate (e.g. neurobiological) and ultimate (evolutionary) aspects of surprise and brings new insights into key aspects of prey defences that exploit predator surprise. We focus on surprise in defences that involve a sudden change in prey appearance (such as deimatic displays), and in defences that increase prey unpredictability (such as variation in chemical defences). We review literature that have investigated these phenomena and connect them to active inference. We also consider how multiple prey defences impact surprise in predators. Finally, we consider the implications of active inference for future studies of predator–prey interactions, illustrate how this approach can be used to quantify surprise in prey defences and predict predator behaviour, and outline key questions that can be addressed within this framework. Read the free Plain Language Summary for this article on the Journal blog.