James Hensman
James Hensman has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Keith Worden
University of Sheffield
12 shared publications - Magnus Rattray
University of Manchester
10 shared publications - Mark J. Eaton
Cardiff School of Engineering Cardiff University Cardiff UK
6 shared publications - Sam L. Evans
School of Engineering, Cardiff University, The Parade, Cardiff CF24 3AA, UK
6 shared publications - Karen M. Holford
Cardiff School of Engineering, Cardiff University, Cardiff, UK
5 shared publications - Cecilia Surace
Politecnico di Torino
3 shared publications
Latest Publications
- Open Access
- 29 May 2018
- Published by Springer Nature in Genome Biology
- Vol. 19 (1) , 65
BGP: identifying gene-specific branching dynamics from single-cell data with a branching Gaussian process
High-throughput single-cell gene expression experiments can be used to uncover branching dynamics in cell populations undergoing differentiation through pseudotime methods. We develop the branching Gaussian process (BGP), a non-parametric model that is able to identify branching dynamics for individual genes and provide an estimate of branching times for each gene with an associated credible region. We demonstrate the effectiveness of our method on simulated data, a single-cell RNA-seq haematopoiesis study and mouse embryonic stem cells generated using droplet barcoding. The method is robust to high levels of technical variation and dropout, which are common in single-cell data.
- 29 May 2018
- Published by Springer Nature in Genome Biology
- Vol. 19 (1) , 65
Open AccessBGP: identifying gene-specific branching dynamics from single-cell data with a branching Gaussian process
High-throughput single-cell gene expression experiments can be used to uncover branching dynamics in cell populations undergoing differentiation through pseudotime methods. We develop the branching Gaussian process (BGP), a non-parametric model that is able to identify branching dynamics for individual genes and provide an estimate of branching times for each gene with an associated credible region. We demonstrate the effectiveness of our method on simulated data, a single-cell RNA-seq haematopoiesis study and mouse embryonic stem cells generated using droplet barcoding. The method is robust to high levels of technical variation and dropout, which are common in single-cell data.
- Open Access
BGP: Branched Gaussian processes for identifying gene-specific branching dynamics in single cell data
High-throughput single-cell gene expression experiments can be used to uncover branching dynamics in cell populations undergoing differentiation through use of pseudotime methods. We develop the branching Gaussian process (BGP), a non-parametric model that is able to identify branching dynamics for individual genes and provides an estimate of branching times for each gene with an associated credible region. We demonstrate the effectiveness of our method on both synthetic data and a published single-cell gene expression hematopoiesis study. The method requires prior information about pseudotime and global cellular branching for each cell but the probabilistic nature of the method means that it is robust to errors in these global branch labels and can be used to discover early branching genes which diverge before the inferred global cell branching. The code is open-source and available at https://github.com/ManchesterBioinference/BranchedGP .
Open AccessBGP: Branched Gaussian processes for identifying gene-specific branching dynamics in single cell data
High-throughput single-cell gene expression experiments can be used to uncover branching dynamics in cell populations undergoing differentiation through use of pseudotime methods. We develop the branching Gaussian process (BGP), a non-parametric model that is able to identify branching dynamics for individual genes and provides an estimate of branching times for each gene with an associated credible region. We demonstrate the effectiveness of our method on both synthetic data and a published single-cell gene expression hematopoiesis study. The method requires prior information about pseudotime and global cellular branching for each cell but the probabilistic nature of the method means that it is robust to errors in these global branch labels and can be used to discover early branching genes which diverge before the inferred global cell branching. The code is open-source and available at https://github.com/ManchesterBioinference/BranchedGP .
- Open Access
- 1 April 2017
- Published by Elsevier in Progress in Aerospace Sciences
- Vol. 90, 1-11
A new methodology for automating acoustic emission detection of metallic fatigue fractures in highly demanding aerospace environments: An overview
- 1 April 2017
- Published by Elsevier in Progress in Aerospace Sciences
- Vol. 90, 1-11
Open AccessA new methodology for automating acoustic emission detection of metallic fatigue fractures in highly demanding aerospace environments: An overview
- Open Access
- 13 April 2016
- Published by PeerJ in PeerJ Computer Science
- Vol. 2, e50
Detecting periodicities with Gaussian processes
We consider the problem of detecting and quantifying the periodic component of a function given noise-corrupted observations of a limited number of input/output tuples. Our approach is based on Gaussian process regression, which provides a flexible non-parametric framework for modelling periodic data. We introduce a novel decomposition of the covariance function as the sum of periodic and aperiodic kernels. This decomposition allows for the creation of sub-models which capture the periodic nature of the signal and its complement. To quantify the periodicity of the signal, we derive a periodicity ratio which reflects the uncertainty in the fitted sub-models. Although the method can be applied to many kernels, we give a special emphasis to the Matérn family, from the expression of the reproducing kernel Hilbert space inner product to the implementation of the associated periodic kernels in a Gaussian process toolkit. The proposed method is illustrated by considering the detection of periodically expressed genes in thearabidopsisgenome.
- 13 April 2016
- Published by PeerJ in PeerJ Computer Science
- Vol. 2, e50
Open AccessDetecting periodicities with Gaussian processes
We consider the problem of detecting and quantifying the periodic component of a function given noise-corrupted observations of a limited number of input/output tuples. Our approach is based on Gaussian process regression, which provides a flexible non-parametric framework for modelling periodic data. We introduce a novel decomposition of the covariance function as the sum of periodic and aperiodic kernels. This decomposition allows for the creation of sub-models which capture the periodic nature of the signal and its complement. To quantify the periodicity of the signal, we derive a periodicity ratio which reflects the uncertainty in the fitted sub-models. Although the method can be applied to many kernels, we give a special emphasis to the Matérn family, from the expression of the reproducing kernel Hilbert space inner product to the implementation of the associated periodic kernels in a Gaussian process toolkit. The proposed method is illustrated by considering the detection of periodically expressed genes in thearabidopsisgenome.
- Open Access
- 26 August 2015
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 31 (24) , 3881-3889
Fast and accurate approximate inference of transcript expression from RNA-seq data
Motivation: Assigning RNA-seq reads to their transcript of origin is a fundamental task in transcript expression estimation. Where ambiguities in assignments exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem can be solved through probabilistic inference. Bayesian methods have been shown to provide accurate transcript abundance estimates compared with competing methods. However, exact Bayesian inference is intractable and approximate methods such as Markov chain Monte Carlo and Variational Bayes (VB) are typically used. While providing a high degree of accuracy and modelling flexibility, standard implementations can be prohibitively slow for large datasets and complex transcriptome annotations. Results: We propose a novel approximate inference scheme based on VB and apply it to an existing model of transcript expression inference from RNA-seq data. Recent advances in VB algorithmics are used to improve the convergence of the algorithm beyond the standard Variational Bayes Expectation Maximization algorithm. We apply our algorithm to simulated and biological datasets, demonstrating a significant increase in speed with only very small loss in accuracy of expression level estimation. We carry out a comparative study against seven popular alternative methods and demonstrate that our new algorithm provides excellent accuracy and inter-replicate consistency while remaining competitive in computation time. Availability and implementation: The methods were implemented in R and C++, and are available as part of the BitSeq project at github.com/BitSeq. The method is also available through the BitSeq Bioconductor package. The source code to reproduce all simulation results can be accessed via github.com/BitSeq/BitSeqVB_benchmarking. Contact:james.hensman@sheffield.ac.uk or panagiotis.papastamoulis@manchester.ac.uk or Magnus.Rattray@manchester.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.
- 26 August 2015
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 31 (24) , 3881-3889
Open AccessFast and accurate approximate inference of transcript expression from RNA-seq data
Motivation: Assigning RNA-seq reads to their transcript of origin is a fundamental task in transcript expression estimation. Where ambiguities in assignments exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem can be solved through probabilistic inference. Bayesian methods have been shown to provide accurate transcript abundance estimates compared with competing methods. However, exact Bayesian inference is intractable and approximate methods such as Markov chain Monte Carlo and Variational Bayes (VB) are typically used. While providing a high degree of accuracy and modelling flexibility, standard implementations can be prohibitively slow for large datasets and complex transcriptome annotations. Results: We propose a novel approximate inference scheme based on VB and apply it to an existing model of transcript expression inference from RNA-seq data. Recent advances in VB algorithmics are used to improve the convergence of the algorithm beyond the standard Variational Bayes Expectation Maximization algorithm. We apply our algorithm to simulated and biological datasets, demonstrating a significant increase in speed with only very small loss in accuracy of expression level estimation. We carry out a comparative study against seven popular alternative methods and demonstrate that our new algorithm provides excellent accuracy and inter-replicate consistency while remaining competitive in computation time. Availability and implementation: The methods were implemented in R and C++, and are available as part of the BitSeq project at github.com/BitSeq. The method is also available through the BitSeq Bioconductor package. The source code to reproduce all simulation results can be accessed via github.com/BitSeq/BitSeqVB_benchmarking. Contact:james.hensman@sheffield.ac.uk or panagiotis.papastamoulis@manchester.ac.uk or Magnus.Rattray@manchester.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.
- Open Access
Fast and accurate approximate inference of transcript expression from RNA-seq data
Motivation: Assigning RNA-seq reads to their transcript of origin is a fundamental task in transcript expression estimation. Where ambiguities in assignments exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem can be solved through probabilistic inference. Bayesian methods have been shown to provide accurate transcript abundance estimates compared to competing methods. However, exact Bayesian inference is intractable and approximate methods such as Markov chain Monte Carlo (MCMC) and Variational Bayes (VB) are typically used. While providing a high degree of accuracy and modelling flexibility, standard implementations can be prohibitively slow for large datasets and complex transcriptome annotations. Results: We propose a novel approximate inference scheme based on VB and apply it to an existing model of transcript expression inference from RNA-seq data. Recent advances in VB algorithmics are used to improve the convergence of the algorithm beyond the standard Variational Bayes Expectation Maximisation (VBEM) algorithm. We apply our algorithm to simulated and biological datasets, demonstrating a significant increase in speed with only very small loss in accuracy of expression level estimation. We carry out a comparative study against seven popular alternative methods and demonstrate that our new algorithm provides excellent accuracy and inter-replicate consistency while remaining competitive in computation time. Availability: The methods were implemented in R and C++, and are available as part of the BitSeq project at \url{https://github.com/BitSeq}. The method is also available through the BitSeq Bioconductor package. The source code to reproduce all simulation results can be accessed via \url{https://github.com/BitSeq/BitSeqVB_benchmarking}.
Open AccessFast and accurate approximate inference of transcript expression from RNA-seq data
Motivation: Assigning RNA-seq reads to their transcript of origin is a fundamental task in transcript expression estimation. Where ambiguities in assignments exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem can be solved through probabilistic inference. Bayesian methods have been shown to provide accurate transcript abundance estimates compared to competing methods. However, exact Bayesian inference is intractable and approximate methods such as Markov chain Monte Carlo (MCMC) and Variational Bayes (VB) are typically used. While providing a high degree of accuracy and modelling flexibility, standard implementations can be prohibitively slow for large datasets and complex transcriptome annotations. Results: We propose a novel approximate inference scheme based on VB and apply it to an existing model of transcript expression inference from RNA-seq data. Recent advances in VB algorithmics are used to improve the convergence of the algorithm beyond the standard Variational Bayes Expectation Maximisation (VBEM) algorithm. We apply our algorithm to simulated and biological datasets, demonstrating a significant increase in speed with only very small loss in accuracy of expression level estimation. We carry out a comparative study against seven popular alternative methods and demonstrate that our new algorithm provides excellent accuracy and inter-replicate consistency while remaining competitive in computation time. Availability: The methods were implemented in R and C++, and are available as part of the BitSeq project at \url{https://github.com/BitSeq}. The method is also available through the BitSeq Bioconductor package. The source code to reproduce all simulation results can be accessed via \url{https://github.com/BitSeq/BitSeqVB_benchmarking}.
- Open Access
MCMC for Variationally Sparse Gaussian Processes
Gaussian process (GP) models form a core part of probabilistic machine learning. Considerable research effort has been made into attacking three issues with GP models: how to compute efficiently when the number of data is large; how to approximate the posterior when the likelihood is not Gaussian and how to estimate covariance function parameter posteriors. This paper simultaneously addresses these, using a variational approximation to the posterior which is sparse in support of the function but otherwise free-form. The result is a Hybrid Monte-Carlo sampling scheme which allows for a non-Gaussian approximation over the function values and covariance parameters simultaneously, with efficient computations based on inducing-point sparse GPs. Code to replicate each experiment in this paper will be available shortly.
Open AccessMCMC for Variationally Sparse Gaussian Processes
Gaussian process (GP) models form a core part of probabilistic machine learning. Considerable research effort has been made into attacking three issues with GP models: how to compute efficiently when the number of data is large; how to approximate the posterior when the likelihood is not Gaussian and how to estimate covariance function parameter posteriors. This paper simultaneously addresses these, using a variational approximation to the posterior which is sparse in support of the function but otherwise free-form. The result is a Hybrid Monte-Carlo sampling scheme which allows for a non-Gaussian approximation over the function values and covariance parameters simultaneously, with efficient computations based on inducing-point sparse GPs. Code to replicate each experiment in this paper will be available shortly.
- Open Access
Spike and Slab Gaussian Process Latent Variable Models
The Gaussian process latent variable model (GP-LVM) is a popular approach to non-linear probabilistic dimensionality reduction. One design choice for the model is the number of latent variables. We present a spike and slab prior for the GP-LVM and propose an efficient variational inference procedure that gives a lower bound of the log marginal likelihood. The new model provides a more principled approach for selecting latent dimensions than the standard way of thresholding the length-scale parameters. The effectiveness of our approach is demonstrated through experiments on real and simulated data. Further, we extend multi-view Gaussian processes that rely on sharing latent dimensions (known as manifold relevance determination) with spike and slab priors. This allows a more principled approach for selecting a subset of the latent space for each view of data. The extended model outperforms the previous state-of-the-art when applied to a cross-modal multimedia retrieval task.
Open AccessSpike and Slab Gaussian Process Latent Variable Models
The Gaussian process latent variable model (GP-LVM) is a popular approach to non-linear probabilistic dimensionality reduction. One design choice for the model is the number of latent variables. We present a spike and slab prior for the GP-LVM and propose an efficient variational inference procedure that gives a lower bound of the log marginal likelihood. The new model provides a more principled approach for selecting latent dimensions than the standard way of thresholding the length-scale parameters. The effectiveness of our approach is demonstrated through experiments on real and simulated data. Further, we extend multi-view Gaussian processes that rely on sharing latent dimensions (known as manifold relevance determination) with spike and slab priors. This allows a more principled approach for selecting a subset of the latent space for each view of data. The extended model outperforms the previous state-of-the-art when applied to a cross-modal multimedia retrieval task.
- 1 May 2015
- Published by Springer Nature in Meccanica
- Vol. 50 (10) , 2537-2546
Damage localisation in delaminated composite plates using a Gaussian process approach
- 1 May 2015
- Published by Springer Nature in Meccanica
- Vol. 50 (10) , 2537-2546
Damage localisation in delaminated composite plates using a Gaussian process approach
- Open Access
- 1 February 2015
- Published by Elsevier in Developmental Cell
- Vol. 32 (3) , 265-277
Hoxa2 Selectively Enhances Meis Binding to Change a Branchial Arch Ground State
- 1 February 2015
- Published by Elsevier in Developmental Cell
- Vol. 32 (3) , 265-277
Open AccessHoxa2 Selectively Enhances Meis Binding to Change a Branchial Arch Ground State