Cailan Jeynes‐Smith

Woodstock and Harris raise concerns about generalisations in our paper ‘Calibrated Ecosystem Models Cannot Predict the Consequences of Conservation Management Decisions’. They implied that our conclusions are invalidated by the exclusion of density dependent compensatory feedback and the existence of automated calibration approaches with overfitting penalties—claims that overlook the broader structural issues we identified. Here, we explore their key criticism in more detail, clarify our position and explain why we continue to stand by our findings.

Ecosystem models are often used to predict the consequences of management interventions in applied ecology and conservation. These models are often high‐dimensional and nonlinear, yet limited data are available to calibrate or validate them. Consequently, their utility as decision‐support tools is unclear. In this paper, we calibrate ecosystem models to time series data from 110 different experimental microcosm ecosystems, each containing three to five interacting species. Then, we assess their ability to predict the consequences of management interventions. Our results show that for each time series dataset, multiple divergent parameter sets offer equivalent, good fits. However, these models have poor predictive accuracy when forecasting future dynamics or when predicting how the ecosystem will respond to management intervention. Closer inspection reveals that the models fail because calibration cannot determine the nature of the interspecific interactions. Our findings question whether ecosystem models can support applied ecological decision‐making when calibrated against real‐world datasets.

Resilient ecological systems are more likely to persist and function in the Anthropocene. Current methods for estimating an ecosystem's resilience rely on accurately parameterized ecosystem models, which is a significant empirical challenge. In this paper, we adapt tools from biochemical kinetics to identify ecological networks that exhibit ‘structural resilience’, a strong form of resilience that is solely a property of the network structure and is independent of model parameters. We undertake an exhaustive search for structural resilience across all three‐species ecological networks, under a generalized Lotka‐Volterra modelling framework. Out of 20,000 possible network structures, approximately 2% display structural resilience. The properties of these networks provide important insights into the mechanisms that could promote resilience in ecosystems, provide new theoretical avenues for qualitative modelling approaches and provide a foundation for identifying robust forms of ecological resilience in large, realistic ecological networks.

Robust perfect adaptation (RPA) is a ubiquitously observed signalling response across all scales of biological organization. A major class of network architectures that drive RPA in complex networks is the Opposer module—a feedback-regulated network into which specialized integral-computing ‘opposer node(s)’ are embedded. Although ultrasensitivity-generating chemical reactions have long been considered a possible mechanism for such adaptation-conferring opposer nodes, this hypothesis has relied on simplified Michaelian models, which neglect the presence of protein–protein complexes. Here we developcomplex-completemodels of interlinked covalent-modification cycles with embedded ultrasensitivity, explicitly capturing all molecular interactions and protein complexes. Strikingly, we demonstrate that the presence of protein–protein complexes thwarts the network’s capacity for RPA in any ‘free’ active protein form, conferring RPA capacity instead on the concentration of a larger protein pool consisting of two distinct forms of a single protein. We further show that the presence of enzyme–substrate complexes, even at comparatively low concentrations, play a crucial and previously unrecognized role in controlling the RPA response—significantly reducing the range of network inputs for which RPA can obtain, and imposing greater parametric requirements on the RPA response. These surprising results raise fundamental new questions as to the biochemical requirements for adaptation-conferring Opposer modules within complex cellular networks.

Robust Perfect Adaptation (RPA) is a ubiquitously-observed signalling response across all scales of biological organisation. A major class of network architectures that drive RPA in complex networks is the Opposer module – a feedback-regulated network into which specialised integral-computing ‘opposer node(s)’ are embedded. Although ultrasensitivity-generating chemical reactions have long been considered a possible mechanism for such adaptation-conferring opposer nodes, this hypothesis has relied on simplified Michaelian models, which neglect the presence of protein-protein complexes, and which are now widely acknowledged to make inaccurate predictions of signalling responses. Here we develop complex-complete models of interlinked covalent-modification cycles with embedded ultrasensitivity: explicitly capturing all molecular interactions and protein complexes. Strikingly, we demonstrate that the presence of protein-protein complexes thwarts the network’s capacity for RPA in any ‘free’ active protein form, conferring RPA capacity instead on the concentration of a larger protein pool consisting of two distinct forms of a single protein. Furthermore, compared to predictions by simplified models, the parametric requirements for RPA in this protein pool are much more severe, and RPA generally obtains over a narrower range of input stimuli. These surprising results raise fundamental new questions as to the biochemical requirements for adaptation-conferring Opposer modules within complex cellular networks.

Switch-like behaviours in biochemical networks are of fundamental significance in biological signal processing, and exist as two distinct types: ultra-sensitivity and bistability. Here we propose two new models of a reversible covalent-modification cycle with positive autoregulation (PAR), a motif structure that is thought to be capable of both ultrasensitivity and bistability in different parameter regimes. These new models appeal to a modelling framework that we call complex-complete , which accounts fully for the molecular complexities of the underlying signalling mechanisms. Each of the two new models encodes a specific molecular mechanism for PAR. We demonstrate that the modelling simplifications for PAR models that have been used in previous work, which rely on Michaelian approximations, are unable to accurately recapitulate the qualitative signalling responses supported by our detailed models. Strikingly, we show that complex-complete PAR models are capable of new qualitative responses such as one-way switches and a ‘prozone’ effect, depending on the specific PAR-encoding mechanism, which are not supported by Michaelian simplifications. Our results highlight the critical importance of accurately representing the molecular details of biochemical signalling mechanisms, and strongly suggest that the Michaelian approximation is inadequate for predictive models of enzyme-mediated chemical reactions with added regulations such as PAR.

Switch-like behaviours in biochemical networks are of fundamental significance in biological signal processing, and exist as two distinct types: ultra-sensitivity and bistability. Here we propose two new models of a reversible covalent-modification cycle with positive autoregulation (PAR) - a motif structure that is thought to be capable of both ultrasensitivity and bistability in different parameter regimes. These new models appeal to a modelling framework that we call complex complete , which accounts fully for the molecular complexities of the underlying signalling mechanisms. Each of the two new models encodes a specific molecular mechanism for PAR. We demonstrate that the modelling simplifications for PAR models that have been used in previous work, which rely on a Michaelian approximation for the enzyme-mediated reactions, are unable to accurately recapitulate the qualitative signalling responses supported by our ‘full’ complex-complete models. Strikingly, we show that the parameter regimes in which ultrasensitivity and bistability obtain in the complex-complete framework contradict the predictions made by the Michaelian simplification. Our results highlight the critical importance of accurately representing the molecular details of biochemical signalling mechanisms, and strongly suggest that the Michaelian approximation may be inadequate for predictive models of enzyme-mediated chemical reactions with added regulations.