James L. Bown

Towards a general theory of biodiversity

The study of patterns in living diversity is driven by the desire to find the universal rules that underlie the organization of ecosystems. The relative abundance distribution, which characterizes the total number and abundance of species in a community, is arguably the most fundamental measure in ecology. Considerable effort has been expended in striving for a general theory that can explain the form of the distribution. Despite this, a mechanistic understanding of the form in terms of physiological and environmental parameters remains elusive. Recently, it has been proposed that space plays a central role in generating the patterns of diversity. Here we show that an understanding of the observed form of the relative abundance distribution requires a consideration of how individuals pack in time. We present a framework for studying the dynamics of communities which generalizes the prevailing species-based approach to one based on individuals that are characterized by their physiological traits. The observed form of the abundance distribution and its dependence on richness and disturbance are reproduced, and can be understood in terms of the trade-off between time to reproduction and fecundity.

Biomass recycling and the origin of phenotype in fungal mycelia

Fungi are one of the most important and widespread components of the biosphere, and are essential for the growth of over 90% of all vascular plants. Although they are a separate kingdom of life, we know relatively little about the origins of their ubiquitous existence. This reflects a wider ignorance arising from their status as indeterminate organisms epitomized by extreme phenotypic plasticity that is essential for survival in complex environments. Here we show that the fungal phenotype may have its origins in the defining characteristic of indeterminate organisms, namely their ability to recycle locally immobilized internal resources into a mobilized form capable of being directed to new internal sinks. We show that phenotype can be modelled as an emergent phenomenon resulting from the interplay between simple local processes governing uptake and remobilization of internal resources, and macroscopic processes associated with their transport. Observed complex growth forms are reproduced and the sensitive dependence of phenotype on environmental context may be understood in terms of nonlinearities associated with regulation of the recycling apparatus.

Modelling interactions in fungi

Indeterminate organisms have received comparatively little attention in theoretical ecology and still there is much to be understood about the origins and consequences of community structure. The fungi comprise an entire kingdom of life and epitomize the indeterminate growth form. While interactions play a significant role in shaping the community structure of indeterminate organisms, to date most of our knowledge relating to fungi comes from observing interaction outcomes between two species in two-dimensional arena experiments. Interactions in the natural environment are more complex and further insight will benefit from a closer integration of theory and experiment. This requires a modelling framework capable of linking genotype and environment to community structure and function. Towards this, we present a theoretical model that replicates observed interaction outcomes between fungal colonies. The hypotheses underlying the model propose that interaction outcome is an emergent consequence of simple and highly localized processes governing rates of uptake and remobilization of resources, the metabolic cost of production of antagonistic compounds and non-localized transport of internal resources. The model may be used to study systems of many interacting colonies and so provides a platform upon which the links between individual-scale behaviour and community-scale function in complex environments can be built.

The role of modelling in identifying drug targets for diseases of the cell cycle

The cell cycle is implicated in diseases that are the leading cause of mortality and morbidity in the developed world. Until recently, the search for drug targets has focused on relatively small parts of the regulatory network under the assumption that key events can be controlled by targeting single pathways. This is valid provided the impact of couplings to the wider scale context of the network can be ignored. The resulting depth of study has revealed many new insights; however, these have been won at the expense of breadth and a proper understanding of the consequences of links between the different parts of the network. Since it is now becoming clear that these early assumptions may not hold and successful treatments are likely to employ drugs that simultaneously target a number of different sites in the regulatory network, it is timely to redress this imbalance. However, the substantial increase in complexity presents new challenges and necessitates parallel theoretical and experimental approaches. We review the current status of theoretical models for the cell cycle in light of these new challenges. Many of the existing approaches are not sufficiently comprehensive to simultaneously incorporate the required extent of couplings. Where more appropriate levels of complexity are incorporated, the models are difficult to link directly to currently available data. Further progress requires a better integration of experiment and theory. New kinds of data are required that are quantitative, have a higher temporal resolution and that allow simultaneous quantitative comparison of the concentration of larger numbers of different proteins. More comprehensive models are required and must accommodate not only substantial uncertainties in the structure and kinetic parameters of the networks, but also high levels of ignorance. The most recent results relating network complexity to robustness of the dynamics provide clues that suggest progress is possible.

Compensatory effects in the PI3K/PTEN/AKT signaling network following receptor tyrosine kinase inhibition

Overcoming de novo and acquired resistance to anticancer drugs that target signaling networks is a formidable challenge for drug design and effective cancer therapy. Understanding the mechanisms by which this resistance arises may offer a route to addressing the insensitivity of signaling networks to drug intervention and restore the efficacy of anticancer therapy. Extending our recent work identifying PTEN as a key regulator of Herceptin sensitivity, we present an integrated theoretical and experimental approach to study the compensatory mechanisms within the PI3K/PTEN/AKT signaling network that afford resistance to receptor tyrosine kinase (RTK) inhibition by anti-HER2 monoclonal antibodies. In a computational model representing the dynamics of the signaling network, we define a single control parameter that encapsulates the balance of activities of the enzymes involved in the PI3K/PTEN/AKT cycle. By varying this control parameter we are able to demonstrate both distinct dynamic regimes of behavior of the signaling network and the transitions between those regimes. We demonstrate resistance, sensitivity, and suppression of RTK signals by the signaling network. Through model analysis we link the sensitivity-to-resistance transition to specific compensatory mechanisms within the signaling network. We study this transition in detail theoretically by variation of activities of PTEN, PI3K, AKT enzymes, and use the results to inform experiments that perturb the signaling network using combinatorial inhibition of RTK, PTEN, and PI3K enzymes in human ovarian carcinoma cell lines. We find good alignment between theoretical predictions and experimental results. We discuss the application of the results to the challenges of hypersensitivity of the signaling network to RTK signals, suppression of drug resistance, and efficacy of drug combinations in anticancer therapy.

Engineering simulations for cancer systems biology

Computer simulation can be used to inform in vivo and in vitro experimentation, enabling rapid, low-cost hypothesis generation and directing experimental design in order to test those hypotheses. In this way, in silico models become a scientific instrument for investigation, and so should be developed to high standards, be carefully calibrated and their findings presented in such that they may be reproduced. Here, we outline a framework that supports developing simulations as scientific instruments, and we select cancer systems biology as an exemplar domain, with a particular focus on cellular signalling models. We consider the challenges of lack of data, incomplete knowledge and modelling in the context of a rapidly changing knowledge base. Our framework comprises a process to clearly separate scientific and engineering concerns in model and simulation development, and an argumentation approach to documenting models for rigorous way of recording assumptions and knowledge gaps. We propose interactive, dynamic visualisation tools to enable the biological community to interact with cellular signalling models directly for experimental design. There is a mismatch in scale between these cellular models and tissue structures that are affected by tumours, and bridging this gap requires substantial computational resource. We present concurrent programming as a technology to link scales without losing important details through model simplification. We discuss the value of combining this technology, interactive visualisation, argumentation and model separation to support development of multi-scale models that represent biologically plausible cells arranged in biologically plausible structures that model cell behaviour, interactions and response to therapeutic interventions.

Feedforward and feedback regulation of the MAPK and PI3K oscillatory circuit in breast cancer

Although the theoretical possibility of oscillations in MAPK signalling has long been described, experimental validation has proven more elusive. In this study we observed oscillations in MAPK and PI3K signalling in breast cancer cells in response to epidermal growth factor receptor-family stimulation. Using systems level analysis with a kinetic model, we demonstrate that receptor amplification, loss of transcriptional feedback, or pathway crosstalk, are responsible for oscillations in MAPK and PI3K signalling. Transcriptional profiling reveals architectural motifs likely to be responsible for feedback control of oscillations. Overexpression of the HER2 oncogene and inhibition of transcriptional feedback increase the amplitude of oscillations and provide experimental validation of the computational findings.

Features of the reversible sensitivity-resistance transition in PI3K/PTEN/AKT signalling network after HER2 inhibition

Systems biology approaches that combine experimental data and theoretical modelling to understand cellular signalling network dynamics offer a useful platform to investigate the mechanisms of resistance to drug interventions and to identify combination drug treatments. Extending our work on modelling the PI3K/PTEN/AKT signalling network (SN), we analyse the sensitivity of the SN output signal, phospho-AKT, to inhibition of HER2 receptor. We model typical aberrations in this SN identified in cancer development and drug resistance: loss of PTEN activity, PI3K and AKT mutations, HER2 overexpression, and overproduction of GSK3β and CK2 kinases controlling PTEN phosphorylation. We show that HER2 inhibition by the monoclonal antibody pertuzumab increases SN sensitivity, both to external signals and to changes in kinetic parameters of the proteins and their expression levels induced by mutations in the SN. This increase in sensitivity arises from the transition of SN functioning from saturation to non-saturation mode in response to HER2 inhibition. PTEN loss or PIK3CA mutation causes resistance to anti-HER2 inhibitor and leads to the restoration of saturation mode in SN functioning with a consequent decrease in SN sensitivity. We suggest that a drug-induced increase in SN sensitivity to internal perturbations, and specifically mutations, causes SN fragility. In particular, the SN is vulnerable to mutations that compensate for drug action and this may result in a sensitivity-to-resistance transition. The combination of HER2 and PI3K inhibition does not sensitise the SN to internal perturbations (mutations) in the PI3K/PTEN/AKT pathway: this combination treatment provides both synergetic inhibition and may prevent the SN from acquired mutations causing drug resistance. Through combination inhibition treatments, we studied the impact of upstream and downstream interventions to suppress resistance to the HER2 inhibitor in the SN with PTEN loss. Comparison of experimental results of PI3K inhibition in the PTEN upstream pathway with PDK1 inhibition in the PTEN downstream pathway shows that upstream inhibition abrogates resistance to pertuzumab more effectively than downstream inhibition. This difference in inhibition effect arises from the compensatory mechanism of an activation loop induced in the downstream pathway by PTEN loss. We highlight that drug target identification for combination anti-cancer therapy needs to account for the mutation effects on the upstream and downstream pathways.

Argument-Driven Validation of Computer Simulations - A Necessity, Rather than an Option

Research based on computer simulations, especially that conducted through agent-based experimentation, is often criticised for not being a reliable source of information - the simulation software can hide errors or flawed designs that inherently bias results. Consequently, the academic community shows both enthusiasm and lack of trust for such approaches. In order to gain confidence is using engineered systems, domains such as Safety Critical Systems employ structured argumentation techniques as means of explicitly relating claims to evidence - in other words, requirements to deliverables. We argue here that structured argumentation should be used in the development and validation process of simulation-driven research. Making use of the Goal Structuring Notation, we provide insights into how more trustworthy outcomes can be obtained through argumentation-driven validation.

Evidence for emergent behaviour in the community-scale dynamics of a fungal microcosm

A stochastic cellular automaton for modelling the dynamics of a two–species fungal microcosm is presented. The state of each cell in the automaton depends on the state of a predefined neighbourhood via a set of conditional probabilities derived from experiments conducted on pairwise combinations of species. The model is tested by detailed comparison with larger–scale experimental microcosms. By employing different hypotheses which relate the pairwise data to the conditional probabilities in the model, the nature of the local and non–local interactions in the community is explored. The hypothesis that the large–scale dynamics are a consequence of independent interactions between species in a local neighbourhood can be excluded at the 5percnt; significance level. The form of the interdependencies is determined and it is shown that the outcome of the interactions at the local neighbourhood–scale depends on the community–scale patterning of individuals. The dynamics of the microcosm are therefore an emergent property of the system of interacting mycelia that cannot be deduced from a study of the components in isolation.