MCMC inference for Markov Jump Processes via the Linear Noise Approximation

Abstract

Bayesian analysis for Markov jump processes is a non-trivial and challenging problem. Although exact inference is theoretically possible, it is computationally demanding thus its applicability is limited to a small class of problems. In this paper we describe the application of Riemann manifold MCMC methods using an approximation to the likelihood of the Markov jump process which is valid when the system modelled is near its thermodynamic limit. The proposed approach is both statistically and computationally efficient while the convergence rate and mixing of the chains allows for fast MCMC inference. The methodology is evaluated using numerical simulations on two problems from chemical kinetics and one from systems biology.

Cited by in corpus (12)

• Approximation and inference methods for stochastic biochemical kinetics - a tutorial review
• Pooling single-cell recordings: Scalable inference through heterogeneous kinetics
• Stochastic modeling of auto-regulatory genetic feedback loops: a review and comparative study
• Delayed acceptance particle MCMC for exact inference in stochastic kinetic models
• Information-geometric Markov Chain Monte Carlo methods using Diffusions
• Quantifying intrinsic and extrinsic noise in gene transcription using the linear noise approximation: An application to single cell data
• Tensor-train approximation of the chemical master equation and its application for parameter inference
• Bayesian Inference for Hybrid Discrete-Continuous Stochastic Kinetic Models
• Accelerating sequential Monte Carlo with surrogate likelihoods
• Parameter inference method for stochastic single-cell dynamics from tree-structured data
• Efficient sampling of conditioned Markov jump processes
• Information and contact geometric description of expectation variables exactly derived from master equations