English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The development of chemical reaction models aids understanding and predictionin areas ranging from biology to electrochemistry and combustion. A systematicapproach to building reaction network models uses observational data not onlyto estimate unknown parameters, but also to learn model structure. Bayesianinference provides a natural approach to this data-driven construction ofmodels. Yet traditional Bayesian model inference methodologies that numericallyevaluate the evidence for each model are often infeasible for nonlinearreaction network inference, as the number of plausible models can becombinatorially large. Alternative approaches based on model-space sampling canenable large-scale network inference, but their realization presents manychallenges. In this paper, we present new computational methods that makelarge-scale nonlinear network inference tractable. First, we exploit thetopology of networks describing potential interactions among chemical speciesto design improved "between-model" proposals for reversible-jump Markov chainMonte Carlo. Second, we introduce a sensitivity-based determination of movetypes which, when combined with network-aware proposals, yields significantadditional gains in sampling performance. These algorithms are demonstrated oninference problems drawn from systems biology, with nonlinear differentialequation models of species interactions.

Exploiting network topology for large-scale inference of nonlinear reaction models