Robert Salomone
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Robert Salomone has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Adrianne L. Jenner
School of Mathematical Sciences, Queensland University of Technology
3 shared publications - Christopher Drovandi
School of Mathematical Sciences Queensland University of Technology Brisbane Australia
3 shared publications - Carl Schmitz
School of Computer Science, Queensland University of Technology
2 shared publications - David James Warne
School of Mathematical Sciences, Queensland University of Technology
2 shared publications - Dimitri Perrin
Queensland University of Technology
2 shared publications - Fred Roosta
The University of Queensland
2 shared publications
Latest Publications
- Open Access
- 15 February 2024
- Published by Springer Nature in Journal of Mathematical Biology
- Vol. 88 (3) , 1-25
Calibration of agent based models for monophasic and biphasic tumour growth using approximate Bayesian computation
Agent-based models (ABMs) are readily used to capture the stochasticity in tumour evolution; however, these models are often challenging to validate with experimental measurements due to model complexity. The Voronoi cell-based model (VCBM) is an off-lattice agent-based model that captures individual cell shapes using a Voronoi tessellation and mimics the evolution of cancer cell proliferation and movement. Evidence suggests tumours can exhibit biphasic growth in vivo. To account for this phenomena, we extend the VCBM to capture the existence of two distinct growth phases. Prior work primarily focused on point estimation for the parameters without consideration of estimating uncertainty. In this paper, approximate Bayesian computation is employed to calibrate the model to in vivo measurements of breast, ovarian and pancreatic cancer. Our approach involves estimating the distribution of parameters that govern cancer cell proliferation and recovering outputs that match the experimental data. Our results show that the VCBM, and its biphasic extension, provides insight into tumour growth and quantifies uncertainty in the switching time between the two phases of the biphasic growth model. We find this approach enables precise estimates for the time taken for a daughter cell to become a mature cell. This allows us to propose future refinements to the model to improve accuracy, whilst also making conclusions about the differences in cancer cell characteristics.
- 15 February 2024
- Published by Springer Nature in Journal of Mathematical Biology
- Vol. 88 (3) , 1-25
Open AccessCalibration of agent based models for monophasic and biphasic tumour growth using approximate Bayesian computation
Agent-based models (ABMs) are readily used to capture the stochasticity in tumour evolution; however, these models are often challenging to validate with experimental measurements due to model complexity. The Voronoi cell-based model (VCBM) is an off-lattice agent-based model that captures individual cell shapes using a Voronoi tessellation and mimics the evolution of cancer cell proliferation and movement. Evidence suggests tumours can exhibit biphasic growth in vivo. To account for this phenomena, we extend the VCBM to capture the existence of two distinct growth phases. Prior work primarily focused on point estimation for the parameters without consideration of estimating uncertainty. In this paper, approximate Bayesian computation is employed to calibrate the model to in vivo measurements of breast, ovarian and pancreatic cancer. Our approach involves estimating the distribution of parameters that govern cancer cell proliferation and recovering outputs that match the experimental data. Our results show that the VCBM, and its biphasic extension, provides insight into tumour growth and quantifies uncertainty in the switching time between the two phases of the biphasic growth model. We find this approach enables precise estimates for the time taken for a daughter cell to become a mature cell. This allows us to propose future refinements to the model to improve accuracy, whilst also making conclusions about the differences in cancer cell characteristics.
- Open Access
Leveraging uncertainty quantification to optimise CRISPR guide RNA selection
CRISPR-based genome editing relies on guide RNA sequences to target specific regions of interest. A large number of methods have been developed to predict how efficient different guides are at inducing indels. As more experimental data becomes available, methods based on machine learning have become more prominent. Here, we explore whether quantifying the uncertainty around these predictions can be used to design better guide selection strategies. We demonstrate how using a deep ensemble approach achieves better performance than utilising a single model. This approach can also provide uncertainty quantification. This allows to design, for the first time, strategies that consider uncertainty in guide RNA selection. These strategies achieve precision over 91% and can identify suitable guides for more than 93% of genes in the mouse genome. Our deep ensemble model is available at https://github.com/bmdslab/CRISPR_DeepEnsemble .
Open AccessLeveraging uncertainty quantification to optimise CRISPR guide RNA selection
CRISPR-based genome editing relies on guide RNA sequences to target specific regions of interest. A large number of methods have been developed to predict how efficient different guides are at inducing indels. As more experimental data becomes available, methods based on machine learning have become more prominent. Here, we explore whether quantifying the uncertainty around these predictions can be used to design better guide selection strategies. We demonstrate how using a deep ensemble approach achieves better performance than utilising a single model. This approach can also provide uncertainty quantification. This allows to design, for the first time, strategies that consider uncertainty in guide RNA selection. These strategies achieve precision over 91% and can identify suitable guides for more than 93% of genes in the mouse genome. Our deep ensemble model is available at https://github.com/bmdslab/CRISPR_DeepEnsemble .
- Open Access
Fast and scalable off-target assessment for CRISPR guide RNAs using partial matches
The design of CRISPR-Cas9 guide RNAs is not trivial. In particular, it is crucial to evaluate the risk of unintended, off-target modifications, but this is computationally expensive. To avoid a brute-force approach where each guide RNA is compared against every possible CRISPR target site in the genome, we previously introduced Crackling, a guide RNA design tool that relies on exact matches over 4bp subsequences to approximate a neighbourhood and accelerate off-target scoring by greatly reducing the search space. While this was faster than other existing tools, it still generates large neighbourhoods. Here, we aim to further reduce the search space by requiring more, now non-contiguous, exact matches. The new implementation, called Crackling++, is benchmarked against our initial approach and other off-target evaluation tools. We show that it provides the fastest way to assess candidate guide RNAs. By using memorymapped files, it also scales to the largest genomes. Crackling++ is available at https://github.com/bmds-lab/CracklingPlusPlus under the Berkeley Software Distribution (BSD) 3-Clause license.
Open AccessFast and scalable off-target assessment for CRISPR guide RNAs using partial matches
The design of CRISPR-Cas9 guide RNAs is not trivial. In particular, it is crucial to evaluate the risk of unintended, off-target modifications, but this is computationally expensive. To avoid a brute-force approach where each guide RNA is compared against every possible CRISPR target site in the genome, we previously introduced Crackling, a guide RNA design tool that relies on exact matches over 4bp subsequences to approximate a neighbourhood and accelerate off-target scoring by greatly reducing the search space. While this was faster than other existing tools, it still generates large neighbourhoods. Here, we aim to further reduce the search space by requiring more, now non-contiguous, exact matches. The new implementation, called Crackling++, is benchmarked against our initial approach and other off-target evaluation tools. We show that it provides the fastest way to assess candidate guide RNAs. By using memorymapped files, it also scales to the largest genomes. Crackling++ is available at https://github.com/bmds-lab/CracklingPlusPlus under the Berkeley Software Distribution (BSD) 3-Clause license.
- Open Access
A PAC-Bayesian Perspective on the Interpolating Information Criterion
Deep learning is renowned for its theory-practice gap, whereby principled theory typically fails to provide much beneficial guidance for implementation in practice. This has been highlighted recently by the benign overfitting phenomenon: when neural networks become sufficiently large to interpolate the dataset perfectly, model performance appears to improve with increasing model size, in apparent contradiction with the well-known bias-variance tradeoff. While such phenomena have proven challenging to theoretically study for general models, the recently proposed Interpolating Information Criterion (IIC) provides a valuable theoretical framework to examine performance for overparameterized models. Using the IIC, a PAC-Bayes bound is obtained for a general class of models, characterizing factors which influence generalization performance in the interpolating regime. From the provided bound, we quantify how the test error for overparameterized models achieving effectively zero training error depends on the quality of the implicit regularization imposed by e.g. the combination of model, optimizer, and parameter-initialization scheme; the spectrum of the empirical neural tangent kernel; curvature of the loss landscape; and noise present in the data.
Open AccessA PAC-Bayesian Perspective on the Interpolating Information Criterion
Deep learning is renowned for its theory-practice gap, whereby principled theory typically fails to provide much beneficial guidance for implementation in practice. This has been highlighted recently by the benign overfitting phenomenon: when neural networks become sufficiently large to interpolate the dataset perfectly, model performance appears to improve with increasing model size, in apparent contradiction with the well-known bias-variance tradeoff. While such phenomena have proven challenging to theoretically study for general models, the recently proposed Interpolating Information Criterion (IIC) provides a valuable theoretical framework to examine performance for overparameterized models. Using the IIC, a PAC-Bayes bound is obtained for a general class of models, characterizing factors which influence generalization performance in the interpolating regime. From the provided bound, we quantify how the test error for overparameterized models achieving effectively zero training error depends on the quality of the implicit regularization imposed by e.g. the combination of model, optimizer, and parameter-initialization scheme; the spectrum of the empirical neural tangent kernel; curvature of the loss landscape; and noise present in the data.
- Open Access
Federated Variational Inference Methods for Structured Latent Variable Models
Federated learning methods enable model training across distributed data sources without data leaving their original locations and have gained increasing interest in various fields. However, existing approaches are limited, excluding many structured probabilistic models. We present a general and elegant solution based on structured variational inference, widely used in Bayesian machine learning, adapted for the federated setting. Additionally, we provide a communication-efficient variant analogous to the canonical FedAvg algorithm. The proposed algorithms' effectiveness is demonstrated, and their performance is compared with hierarchical Bayesian neural networks and topic models.
Open AccessFederated Variational Inference Methods for Structured Latent Variable Models
Federated learning methods enable model training across distributed data sources without data leaving their original locations and have gained increasing interest in various fields. However, existing approaches are limited, excluding many structured probabilistic models. We present a general and elegant solution based on structured variational inference, widely used in Bayesian machine learning, adapted for the federated setting. Additionally, we provide a communication-efficient variant analogous to the canonical FedAvg algorithm. The proposed algorithms' effectiveness is demonstrated, and their performance is compared with hierarchical Bayesian neural networks and topic models.
- Open Access
Being Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics
Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products.
Open AccessBeing Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics
Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products.
- Open Access
- 13 October 2022
- Published by Elsevier in Econometrics and Statistics
- Vol. 32, 98-121
Spectral Subsampling MCMC for Stationary Multivariate Time Series with Applications to Vector ARTFIMA Processes
- 13 October 2022
- Published by Elsevier in Econometrics and Statistics
- Vol. 32, 98-121
Open AccessSpectral Subsampling MCMC for Stationary Multivariate Time Series with Applications to Vector ARTFIMA Processes
- Open Access
Calibration of Agent Based Models for Monophasic and Biphasic Tumour Growth using Approximate Bayesian Computation
Agent-based models (ABMs) are readily used to capture the stochasticity in tumour evolution; however, these models are often challenging to validate with experimental measurements due to model complexity. The Voronoi cell-based model (VCBM) is an off-lattice agent-based model that captures individual cell shapes using a Voronoi tessellation and mimics the evolution of cancer cell proliferation and movement. Evidence suggests tumours can exhibit biphasic growth in vivo . To account for this phenomena, we extend the VCBM to capture the existence of two distinct growth phases. Prior work primarily focused on point estimation for the parameters without consideration of estimating uncertainty. In this paper, approximate Bayesian computation is employed to calibrate the model to in vivo measurements of breast, ovarian and pancreatic cancer. Our approach involves estimating the distribution of parameters that govern cancer cell proliferation and recovering outputs that match the experimental data. Our results show that the VCBM, and its biphasic extension, provides insight into tumour growth and quantifies uncertainty in the switching time between the two phases of the biphasic growth model. We find this approach enables precise estimates for the time taken for a daughter cell to become a mature cell. This allows us to propose future refinements to the model to improve accuracy, whilst also making conclusions about the differences in cancer cell characteristics.
Open AccessCalibration of Agent Based Models for Monophasic and Biphasic Tumour Growth using Approximate Bayesian Computation
Agent-based models (ABMs) are readily used to capture the stochasticity in tumour evolution; however, these models are often challenging to validate with experimental measurements due to model complexity. The Voronoi cell-based model (VCBM) is an off-lattice agent-based model that captures individual cell shapes using a Voronoi tessellation and mimics the evolution of cancer cell proliferation and movement. Evidence suggests tumours can exhibit biphasic growth in vivo . To account for this phenomena, we extend the VCBM to capture the existence of two distinct growth phases. Prior work primarily focused on point estimation for the parameters without consideration of estimating uncertainty. In this paper, approximate Bayesian computation is employed to calibrate the model to in vivo measurements of breast, ovarian and pancreatic cancer. Our approach involves estimating the distribution of parameters that govern cancer cell proliferation and recovering outputs that match the experimental data. Our results show that the VCBM, and its biphasic extension, provides insight into tumour growth and quantifies uncertainty in the switching time between the two phases of the biphasic growth model. We find this approach enables precise estimates for the time taken for a daughter cell to become a mature cell. This allows us to propose future refinements to the model to improve accuracy, whilst also making conclusions about the differences in cancer cell characteristics.
- Open Access
Continuously-Tempered PDMP Samplers
New sampling algorithms based on simulating continuous-time stochastic processes called piece-wise deterministic Markov processes (PDMPs) have shown considerable promise. However, these methods can struggle to sample from multi-modal or heavy-tailed distributions. We show how tempering ideas can improve the mixing of PDMPs in such cases. We introduce an extended distribution defined over the state of the posterior distribution and an inverse temperature, which interpolates between a tractable distribution when the inverse temperature is 0 and the posterior when the inverse temperature is 1. The marginal distribution of the inverse temperature is a mixture of a continuous distribution on [0,1) and a point mass at 1: which means that we obtain samples when the inverse temperature is 1, and these are draws from the posterior, but sampling algorithms will also explore distributions at lower temperatures which will improve mixing. We show how PDMPs, and particularly the Zig-Zag sampler, can be implemented to sample from such an extended distribution. The resulting algorithm is easy to implement and we show empirically that it can outperform existing PDMP-based samplers on challenging multimodal posteriors.
Open AccessContinuously-Tempered PDMP Samplers
New sampling algorithms based on simulating continuous-time stochastic processes called piece-wise deterministic Markov processes (PDMPs) have shown considerable promise. However, these methods can struggle to sample from multi-modal or heavy-tailed distributions. We show how tempering ideas can improve the mixing of PDMPs in such cases. We introduce an extended distribution defined over the state of the posterior distribution and an inverse temperature, which interpolates between a tractable distribution when the inverse temperature is 0 and the posterior when the inverse temperature is 1. The marginal distribution of the inverse temperature is a mixture of a continuous distribution on [0,1) and a point mass at 1: which means that we obtain samples when the inverse temperature is 1, and these are draws from the posterior, but sampling algorithms will also explore distributions at lower temperatures which will improve mixing. We show how PDMPs, and particularly the Zig-Zag sampler, can be implemented to sample from such an extended distribution. The resulting algorithm is easy to implement and we show empirically that it can outperform existing PDMP-based samplers on challenging multimodal posteriors.
- Open Access
Implicit Langevin Algorithms for Sampling From Log-concave Densities
For sampling from a log-concave density, we study implicit integrators resulting from $\theta$-method discretization of the overdamped Langevin diffusion stochastic differential equation. Theoretical and algorithmic properties of the resulting sampling methods for $ \theta \in [0,1] $ and a range of step sizes are established. Our results generalize and extend prior works in several directions. In particular, for $\theta\ge1/2$, we prove geometric ergodicity and stability of the resulting methods for all step sizes. We show that obtaining subsequent samples amounts to solving a strongly-convex optimization problem, which is readily achievable using one of numerous existing methods. Numerical examples supporting our theoretical analysis are also presented.Open AccessImplicit Langevin Algorithms for Sampling From Log-concave Densities
For sampling from a log-concave density, we study implicit integrators resulting from $\theta$-method discretization of the overdamped Langevin diffusion stochastic differential equation. Theoretical and algorithmic properties of the resulting sampling methods for $ \theta \in [0,1] $ and a range of step sizes are established. Our results generalize and extend prior works in several directions. In particular, for $\theta\ge1/2$, we prove geometric ergodicity and stability of the resulting methods for all step sizes. We show that obtaining subsequent samples amounts to solving a strongly-convex optimization problem, which is readily achievable using one of numerous existing methods. Numerical examples supporting our theoretical analysis are also presented.