arXiv subjects

Sylvia Richardson

Publications and source records attributed to Sylvia Richardson.

At least 19 recordsLinked to original sources

arXiv2026

Variational Consensus Monte Carlo for Bayesian Mixture

Motivated by the privacy, sensitivity and sharing limitations of health data, we present a comprehensive pipeline for inference of Bayesian mixture models within a federated learning setting, i.e. when data cannot be fully shared or pooled across compute nodes. We adopt a Consensus Monte Carlo (CMC) approach, in which an MCMC algorithm is run independently within each data silo to estimate local posterior distributions, which are then aggregated to approximate the posterior over the full data. The variational CMC approach of Rabinovich, Angelino and Jordan (2015) [1] frames the aggregation step as a variational inference problem, but their application to mixtures assumes the number of clusters and key mixture parameters to be known. Our main methodological contributions are: (i) an extension of variational CMC to over-fitted Bayesian mixture models that infer the number of clusters and all model parameters, without requiring conjugacy; (ii) novel cluster-matching algorithms suitable for cross-silo settings in which not every cluster appears in each local dataset; (iii) a number of inference strategies for the aggregation step, matched to different federated learning constraints; and (iv) guidelines for choosing among these in practice. A comprehensive simulation study validates the framework and allows us to compare to state-of-the-art federated learning alternatives. Notably, we show that when the composition of local datasets reflects the underlying clustering structure in the data, our approach can recover small clusters with greater accuracy than standard MCMC applied to the pooled data. We illustrate the framework on large-scale electronic health record data, identifying multi-morbidity patterns in a British geriatric population.

stat.ML↗

arXiv2026

Adaptive Bayesian computation for efficient biobank-scale genomic inference

Motivation: Modern biobanks, with unprecedented sample sizes and phenotypic diversity, have become foundational resources for genomic studies, enabling powerful cross-phenotype and population-scale analyses. As studies grow in complexity, Bayesian hierarchical models offer a principled framework for jointly modeling multiple units such as cells, traits, and experimental conditions, increasing statistical power through information sharing. However, adoption of Bayesian hierarchical models in biobank-scale studies remains limited due to computational inefficiencies, particularly in posterior inference over high-dimensional parameter spaces. Deterministic approximations such as variational inference provide scalable alternatives to Markov Chain Monte Carlo, yet current implementations do not fully exploit the structure of genome-wide multi-unit modeling, especially when biological effects of interest are concentrated in a few units. Results: We propose an adaptive focus (AF) strategy within a block coordinate ascent variational inference (CAVI) framework that selectively updates subsets of parameters at each iteration, corresponding to units deemed relevant based on current estimates. We illustrate this approach in protein quantitative trait locus (pQTL) mapping using a joint model of hierarchically linked regressions with shared parameters across traits. In both simulated data and real proteomic data from the UK Biobank, AF-CAVI achieves up to a 50\% reduction in runtime while maintaining statistical performance. We also provide a genome-wide pipeline for multi-trait pQTL mapping across thousands of traits, demonstrating AF-CAVI as an efficient scheme for large-scale, multi-unit Bayesian analysis in biobanks.

stat.AP↗

arXiv2025

A discomfort-informed adaptive Gibbs sampler for finite mixture models

Finite mixture models are frequently used to uncover latent structures in high-dimensional datasets (e.g.\ identifying clusters of patients in electronic health records). The inference of such structures can be performed in a Bayesian framework, and involves the use of sampling algorithms such as Gibbs samplers aimed at deriving posterior distribution of the probabilities of observations to belong to specific clusters. Unfortunately, traditional implementations of Gibbs samplers in this context often face critical challenges, such as inefficient use of computational resources and unnecessary updates for observations that are highly likely to remain in their current cluster. This paper introduces a new adaptive Gibbs sampler that improves the convergence efficiency over existing methods. In particular, our sampler is guided by a function that, at each iteration, uses the past of the chain to focus the updating on observations potentially misclassified in the current clustering, i.e.\ those with a low probability of belonging to their current component. Through simulation studies and two real data analyses, we empirically demonstrate that, in terms of convergence time, our method tends to perform more efficiently compared to state-of-the-art approaches.

stat.ME↗

arXiv2025

Federated Variational Inference for Bayesian Mixture Models

We present a federated learning approach for Bayesian model-based clustering of large-scale binary and categorical datasets. We introduce a principled 'divide and conquer' inference procedure using variational inference with local merge and delete moves within batches of the data in parallel, followed by 'global' merge moves across batches to find global clustering structures. We show that these merge moves require only summaries of the data in each batch, enabling federated learning across local nodes without requiring the full dataset to be shared. Empirical results on simulated and benchmark datasets demonstrate that our method performs well in comparison to existing clustering algorithms. We validate the practical utility of the method by applying it to large scale electronic health record (EHR) data.

stat.ML↗

arXiv2024

Efficient Bayesian functional principal component analysis of irregularly-observed multivariate curves

The analysis of multivariate functional curves has the potential to yield important scientific discoveries in domains such as healthcare, medicine, economics and social sciences. However, it is common for real-world settings to present longitudinal data that are both irregularly and sparsely observed, which introduces important challenges for the current functional data methodology. A Bayesian hierarchical framework for multivariate functional principal component analysis is proposed, which accommodates the intricacies of such irregular observation settings by flexibly pooling information across subjects and correlated curves. The model represents common latent dynamics via shared functional principal component scores, thereby effectively borrowing strength across curves while circumventing the computationally challenging task of estimating covariance matrices. These scores also provide a parsimonious representation of the major modes of joint variation of the curves and constitute interpretable scalar summaries that can be employed in follow-up analyses. Estimation is carried out using variational inference, which combines efficiency, modularity and approximate posterior density estimation, enabling the joint analysis of large datasets with parameter uncertainty quantification. Detailed simulations assess the effectiveness of the approach in sharing information from sparse and irregularly sampled multivariate curves. The methodology is also exploited to estimate the molecular disease courses of individual patients with SARS-CoV-2 infection and characterise patient heterogeneity in recovery outcomes; this study reveals key coordinated dynamics across the immune, inflammatory and metabolic systems, which are associated with survival and long-COVID symptoms up to one year post disease onset. The approach is implemented in the R package bayesFPCA.

stat.ME↗

arXiv2023

A large-scale and PCR-referenced vocal audio dataset for COVID-19

The UK COVID-19 Vocal Audio Dataset is designed for the training and evaluation of machine learning models that classify SARS-CoV-2 infection status or associated respiratory symptoms using vocal audio. The UK Health Security Agency recruited voluntary participants through the national Test and Trace programme and the REACT-1 survey in England from March 2021 to March 2022, during dominant transmission of the Alpha and Delta SARS-CoV-2 variants and some Omicron variant sublineages. Audio recordings of volitional coughs, exhalations, and speech were collected in the 'Speak up to help beat coronavirus' digital survey alongside demographic, self-reported symptom and respiratory condition data, and linked to SARS-CoV-2 test results. The UK COVID-19 Vocal Audio Dataset represents the largest collection of SARS-CoV-2 PCR-referenced audio recordings to date. PCR results were linked to 70,794 of 72,999 participants and 24,155 of 25,776 positive cases. Respiratory symptoms were reported by 45.62% of participants. This dataset has additional potential uses for bioacoustics research, with 11.30% participants reporting asthma, and 27.20% with linked influenza PCR test results.

cs.SD↗

arXiv2023

Exploring the big data paradox for various estimands using vaccination data from the global COVID-19 Trends and Impact Survey (CTIS)

Selection bias poses a challenge to statistical inference validity in non-probability surveys. This study compared estimates of the first-dose COVID-19 vaccination rates among Indian adults in 2021 from a large non-probability survey, COVID-19 Trends and Impact Survey (CTIS), and a small probability survey, the Center for Voting Options and Trends in Election Research (CVoter), against benchmark data from the COVID Vaccine Intelligence Network (CoWIN). Notably, CTIS exhibits a larger estimation error (0.39) compared to CVoter (0.16). Additionally, we investigated the estimation accuracy of the CTIS when using a relative scale and found a significant increase in the effective sample size by altering the estimand from the overall vaccination rate. These results suggest that the big data paradox can manifest in countries beyond the US and it may not apply to every estimand of interest.

stat.AP↗

arXiv2023

StabJGL: a stability approach to sparsity and similarity selection in multiple network reconstruction

In recent years, network models have gained prominence for their ability to capture complex associations. In statistical omics, networks can be used to model and study the functional relationships between genes, proteins, and other types of omics data. If a Gaussian graphical model is assumed, a gene association network can be determined from the non-zero entries of the inverse covariance matrix of the data. Due to the high-dimensional nature of such problems, integrative methods that leverage similarities between multiple graphical structures have become increasingly popular. The joint graphical lasso is a powerful tool for this purpose, however, the current AIC-based selection criterion used to tune the network sparsities and similarities leads to poor performance in high-dimensional settings. We propose stabJGL, which equips the joint graphical lasso with a stable and accurate penalty parameter selection approach that combines the notion of model stability with likelihood-based similarity selection. The resulting method makes the powerful joint graphical lasso available for use in omics settings, and outperforms the standard joint graphical lasso, as well as state-of-the-art joint methods, in terms of all performance measures we consider. Applying stabJGL to proteomic data from a pan-cancer study, we demonstrate the potential for novel discoveries the method brings. A user-friendly R package for stabJGL with tutorials is available on Github at https://github.com/Camiling/stabJGL.

stat.ME↗

arXiv2023

Scalable Multiple Network Inference with the Joint Graphical Horseshoe

Network models are useful tools for modelling complex associations. If a Gaussian graphical model is assumed, conditional independence is determined by the non-zero entries of the inverse covariance (precision) matrix of the data. The Bayesian graphical horseshoe estimator provides a robust and flexible framework for precision matrix inference, as it introduces local, edge-specific parameters which prevent over-shrinkage of non-zero off-diagonal elements. However, for many applications such as statistical omics, the current implementation based on Gibbs sampling becomes computationally inefficient or even unfeasible in high dimensions. Moreover, the graphical horseshoe has only been formulated for a single network, whereas interest has grown in the network analysis of multiple data sets that might share common structures. We propose (i) a scalable expectation conditional maximisation (ECM) algorithm for obtaining the posterior mode of the precision matrix in the graphical horseshoe, and (ii) a novel joint graphical horseshoe estimator, which borrows information across multiple related networks to improve estimation. We show, on both simulated and real omics data, that our single-network ECM approach is more scalable than the existing graphical horseshoe Gibbs implementation, while achieving the same level of accuracy. We also show that our joint-network proposal successfully leverages shared edge-specific information between networks while still retaining differences, outperforming state-of-the-art methods at any level of network similarity.

stat.ME↗

arXiv2023

Audio-based AI classifiers show no evidence of improved COVID-19 screening over simple symptoms checkers

Recent work has reported that AI classifiers trained on audio recordings can accurately predict severe acute respiratory syndrome coronavirus 2 (SARSCoV2) infection status. Here, we undertake a large scale study of audio-based deep learning classifiers, as part of the UK governments pandemic response. We collect and analyse a dataset of audio recordings from 67,842 individuals with linked metadata, including reverse transcription polymerase chain reaction (PCR) test outcomes, of whom 23,514 tested positive for SARS CoV 2. Subjects were recruited via the UK governments National Health Service Test-and-Trace programme and the REal-time Assessment of Community Transmission (REACT) randomised surveillance survey. In an unadjusted analysis of our dataset AI classifiers predict SARS-CoV-2 infection status with high accuracy (Receiver Operating Characteristic Area Under the Curve (ROCAUC) 0.846 [0.838, 0.854]) consistent with the findings of previous studies. However, after matching on measured confounders, such as age, gender, and self reported symptoms, our classifiers performance is much weaker (ROC-AUC 0.619 [0.594, 0.644]). Upon quantifying the utility of audio based classifiers in practical settings, we find them to be outperformed by simple predictive scores based on user reported symptoms.

cs.SD↗

arXiv2023

Bayesian outcome-guided multi-view mixture models with applications in molecular precision medicine

Clustering is commonly performed as an initial analysis step for uncovering structure in 'omics datasets, e.g. to discover molecular subtypes of disease. The high-throughput, high-dimensional nature of these datasets means that they provide information on a diverse array of different biomolecular processes and pathways. Different groups of variables (e.g. genes or proteins) will be implicated in different biomolecular processes, and hence undertaking analyses that are limited to identifying just a single clustering partition of the whole dataset is therefore liable to conflate the multiple clustering structures that may arise from these distinct processes. To address this, we propose a multi-view Bayesian mixture model that identifies groups of variables (``views"), each of which defines a distinct clustering structure. We consider applications in stratified medicine, for which our principal goal is to identify clusters of patients that define distinct, clinically actionable disease subtypes. We adopt the semi-supervised, outcome-guided mixture modelling approach of Bayesian profile regression that makes use of a response variable in order to guide inference toward the clusterings that are most relevant in a stratified medicine context. We present the model, together with illustrative simulation examples, and examples from pan-cancer proteomics. We demonstrate how the approach can be used to perform integrative clustering, and consider an example in which different 'omics datasets are integrated in the context of breast cancer subtyping.

stat.ME↗

arXiv2022

Distributed Computation for Marginal Likelihood based Model Choice

We propose a general method for distributed Bayesian model choice, using the marginal likelihood, where a data set is split in non-overlapping subsets. These subsets are only accessed locally by individual workers and no data is shared between the workers. We approximate the model evidence for the full data set through Monte Carlo sampling from the posterior on every subset generating a model evidence per subset. The results are combined using a novel approach which corrects for the splitting using summary statistics of the generated samples. Our divide-and-conquer approach enables Bayesian model choice in the large data setting, exploiting all available information but limiting communication between workers. We derive theoretical error bounds that quantify the resulting trade-off between computational gain and loss in precision. The embarrassingly parallel nature yields important speed-ups when used on massive data sets as illustrated by our real world experiments. In addition, we show how the suggested approach can be extended to model choice within a reversible jump setting that explores multiple feature combinations within one run.

stat.CO↗

arXiv2022

Structural randomised selection

An important problem in the analysis of high-dimensional omics data is to identify subsets of molecular variables that are associated with a phenotype of interest. This requires addressing the challenges of high dimensionality, strong multicollinearity and model uncertainty. We propose a new ensemble learning approach for improving the performance of sparse penalised regression methods, called STructural RANDomised Selection (STRANDS). The approach, that builds and improves upon the Random Lasso method, consists of two steps. In both steps, we reduce dimensionality by repeated subsampling of variables. We apply a penalised regression method to each subsampled dataset and average the results. In the first step, subsampling is informed by variable correlation structure, and in the second step, by variable importance measures from the first step. STRANDS can be used with any sparse penalised regression approach as the "base learner". Using synthetic data and real biological datasets, we demonstrate that STRANDS typically improves upon its base learner, and that taking account of the correlation structure in the first step can help to improve the efficiency with which the model space may be explored.

stat.ME↗

arXiv2022

On randomized sketching algorithms and the Tracy-Widom law

There is an increasing body of work exploring the integration of random projection into algorithms for numerical linear algebra. The primary motivation is to reduce the overall computational cost of processing large datasets. A suitably chosen random projection can be used to embed the original dataset in a lower-dimensional space such that key properties of the original dataset are retained. These algorithms are often referred to as sketching algorithms, as the projected dataset can be used as a compressed representation of the full dataset. We show that random matrix theory, in particular the Tracy-Widom law, is useful for describing the operating characteristics of sketching algorithms in the tall-data regime when $n \gg d$. Asymptotic large sample results are of particular interest as this is the regime where sketching is most useful for data compression. In particular, we develop asymptotic approximations for the success rate in generating random subspace embeddings and the convergence probability of iterative sketching algorithms. We test a number of sketching algorithms on real large high-dimensional datasets and find that the asymptotic expressions give accurate predictions of the empirical performance.

math.NA↗

arXiv2021

Bayesian profile regression for clustering analysis involving a longitudinal response and explanatory variables

The identification of sets of co-regulated genes that share a common function is a key question of modern genomics. Bayesian profile regression is a semi-supervised mixture modelling approach that makes use of a response to guide inference toward relevant clusterings. Previous applications of profile regression have considered univariate continuous, categorical, and count outcomes. In this work, we extend Bayesian profile regression to cases where the outcome is longitudinal (or multivariate continuous) and provide PReMiuMlongi, an updated version of PReMiuM, the R package for profile regression. We consider multivariate normal and Gaussian process regression response models and provide proof of principle applications to four simulation studies. The model is applied on budding yeast data to identify groups of genes co-regulated during the Saccharomyces cerevisiae cell cycle. We identify 4 distinct groups of genes associated with specific patterns of gene expression trajectories, along with the bound transcriptional factors, likely involved in their co-regulation process.

stat.ME↗

arXiv2021

Interoperability of statistical models in pandemic preparedness: principles and reality

We present "interoperability" as a guiding framework for statistical modelling to assist policy makers asking multiple questions using diverse datasets in the face of an evolving pandemic response. Interoperability provides an important set of principles for future pandemic preparedness, through the joint design and deployment of adaptable systems of statistical models for disease surveillance using probabilistic reasoning. We illustrate this through case studies for inferring spatial-temporal coronavirus disease 2019 (COVID-19) prevalence and reproduction numbers in England.

stat.ME↗

arXiv2021

BayesSUR: An R package for high-dimensional multivariate Bayesian variable and covariance selection in linear regression

In molecular biology, advances in high-throughput technologies have made it possible to study complex multivariate phenotypes and their simultaneous associations with high-dimensional genomic and other omics data, a problem that can be studied with high-dimensional multi-response regression, where the response variables are potentially highly correlated. To this purpose, we recently introduced several multivariate Bayesian variable and covariance selection models, e.g., Bayesian estimation methods for sparse seemingly unrelated regression for variable and covariance selection. Several variable selection priors have been implemented in this context, in particular the hotspot detection prior for latent variable inclusion indicators, which results in sparse variable selection for associations between predictors and multiple phenotypes. We also propose an alternative, which uses a Markov random field (MRF) prior for incorporating prior knowledge about the dependence structure of the inclusion indicators. Inference of Bayesian seemingly unrelated regression (SUR) by Markov chain Monte Carlo methods is made computationally feasible by factorisation of the covariance matrix amongst the response variables. In this paper we present BayesSUR, an R package, which allows the user to easily specify and run a range of different Bayesian SUR models, which have been implemented in C++ for computational efficiency. The R package allows the specification of the models in a modular way, where the user chooses the priors for variable selection and for covariance selection separately. We demonstrate the performance of sparse SUR models with the hotspot prior and spike-and-slab MRF prior on synthetic and real data sets representing eQTL or mQTL studies and in vitro anti-cancer drug screening studies as examples for typical applications.

stat.ME↗

arXiv2021

The `how' matters: A simulation-based assessment of the potential contributions of LFD tests for school reopening in England

During Covid-19 outbreaks, school closures are employed as part of governments' non-pharmaceutical interventions around the world to reduce the number of contacts and keep the reproduction number below 1. Yet, prolonged school closures have profound negative impact on the future opportunities of pupils, particularly from disadvantaged backgrounds, as well as additional economic and social impacts by preventing their parents from returning to work. Data on Covid-19 in children are sparse and policy frameworks are evolving quickly. We compare a set of potential policies to accompany the reopening of schools by means of an agent-based simulation tool. The policies and scenarios we model reflect the public discussion and government guidelines in early March 2021 in England before the planned nationwide reopening of schools on the 8th of March. A point of particular interest is the potential contribution of a more wide-spread use of screening tests based on lateral flow devices. We compare policies both with respect to their potential to contain new outbreaks of Covid-19 in schools and the proportion of schooldays lost due to isolation of pupils. We find that regular asymptomatic screening of the whole school as an addition to a policy built around isolation of symptomatic pupils and their closest contacts is beneficial across a wide range of scenarios, including when screening tests with relatively low test sensitivity are used. Multiple screening tests per week bring only small additional benefits in some scenarios. These findings remain valid when test compliance is not enforced although the effectiveness of outbreak control is reduced.

stat.AP↗