Talk abstracts

Prof. Damien Lacroix (Bio)
Damien Lacroix is Professor of Biomedical Engineering in the Department of Mechanical Engineering at the University of Sheffield and is Director of Research at the Insigneo Institute for in silico Medicine. His main expertise is on the computational modelling of mechanobiological processes at cell, tissue and organ interfaces. Lacroix has received around £7M in the last 5 years in European and EPSRC funding. He coordinated the only world-widemulti scalepatient specific mechanobiological project focused on the lumbar spine (MySpine). He is also the recipient of a European Research Council (ERC) on multi-scale simulations on bone tissue engineering. He is the PI of an ESPRC Frontier Engineering Award on the Individualised Multi-scale Simulation of the Musculoskeletal System. As past-President of the European Society of Biomechanics (2010- 2012), Lacroix is a leading figure in biomedical engineering.

Multiscale modelling of the musculoskeletal system

Engineering problems are increasing in complexity due to the need to account for (1) multiphysics behaviour where different physical processes are interacting among each other and (2) the inability to describe completely a process at a single space-time scale only and therefore the need to account for interactions among space and time scales. In addition, it is not always possible to measure properties at all those space and time scales so although there is a need to go across scales to solve grand challenges, there is also a need to be able to deal with missing data. Life science is a good example where uncertainty is present everywhere and where deterministic methods are becoming more and more limited. Therefore, there is a need to develop methodologies that include uncertainty. Such challenges are becoming relatively common in many different engineering sectors and therefore the topic of this research is timely.

The use of computer simulations for the development of medical devices or for their use as a pre-clinical tool is novel and the subject of research of a MultiSim EPSRC Frontier Engineering Award awarded the University of Sheffield. MultiSim aims to develop computational models that can simulate musculoskeletal pathologies across scales and under the assumptions that there might be missing or uncertain data or conditions. The ambition of the grant is to show for the first time how the development of new computational tools that integrate multi-scale modelling, unobservable states and variable, and uncertainty can be used in a clinically relevant context for the better understanding or the personalised treatment of some musculoskeletal diseases. In this presentation, an example of the multi-scale approach developed in the lower limb for the prediction of femoral head bone fracture in a patient-specific manner will be described. The conceptual workflow to perform multi-scale simulations will be introduced.


Dr Alex Fletcher (Bio)
Alexander Fletcher is a Vice-Chancellor’s Research Fellow in the School of Mathematics and Statistics at the University of Sheffield. Before moving to Sheffield, he carried out postdoctoral research at the University of Oxford. He develops mathematical models to study biological systems, in close collaboration with experimentalists, with applications to development and cancer. To support this work, since 2005 he has been a lead developer of Chaste, an open-source software library for the simulation of agent-based and multiscale models (

Multiscale modelling of cell populations in a consistent computational framework: progress and challenges

Alongside experimental and clinical approaches, computational modelling offers a useful tool with which to unravel the complex interactions between processes at the intracellular, cellular and tissue scales that underlie the growth of tumours and their response to treatment. A variety of different individual cell-based and multiscale modelling approaches have recently been developed for studying how processes at the level of a single cell affect tissue-level behaviour. While there are clear strengths and weaknesses with each modelling approach, and therefore cases in which it is clear which approach is valid, there are other cases in which it is not clear. When comparing different constitutive assumptions, a computational framework is required that allows one to easily change the fine details of a model and its implementation. A further barrier to the wider use of such models is the lack of standards or benchmarks; models and methods are often not reused effectively, because they are typically not available as rigorously tested, open-source simulation software. It is therefore difficult to guarantee the reproducibility of computational results. In this talk I describe our work to address these issues through the C++ library, Chaste ( I discuss its functionality and the approach we have taken to develop this code, and highlight some of the ongoing computational challenges associated with improving the reproducibility and re-use of individual cell-based and multiscale models.


Prof. Alison Heppenstall (Bio)
Alison Heppenstall is a Professor in Geocomputation in the School of Geography at the University of Leeds. Her interests are in the application of artificial intelligent solutions for geographical problems, with applied research experience in the development and linkage of novel methodologies for a variety of socio-economic applications including cities, education planning/management, retail analysis and crime. A focus of her work is in individual based modelling, in particular the development and application of agent-based modelling, microsimulation and the incorporation of behavioural frameworks.
, Dr Nick Malleson (Bio)
Nick Malleson is an Associate Professor in Geographical Information Science at the Centre for Spatial Analysis and Policy at the School of Geography, University of Leeds, UK. He has a PhD in Geography and undergraduate degrees in Computer Science (BSc) and Multidiciplinary Informatics (MSc). Most of his research focuses on the development of computer models that help us to understand social phenomena, with particular interests in agent-based crime models and simulations of urban dynamics. He has recently been awarded funding from the ESRC (Future Research Leaders) and ERC (Starting Grant) to explore the means of calibrating agent-based models using dynamic streams of data.

Who, why and when? Simulating the Individual in a (big) data-intensive world

Within the last few years we have witnessed a massive increase in computational processing power and storage, combined with the onset of ‘Big Data’. These new data sources (e.g., social media, mobile phone data, etc.) tell us who, why and when, individuals are using spaces. For the agent-based modeler, this is an unprecedented opportunity to build robust models that can both simulate the individual in detail and provide new understanding about the consequences of individual decisions across space and time. We will present current and planned work from the newly formed Agent-Based Modelling centre at Leeds that are focused on answering the following research questions: can we effectively use novel data to understand and model spatial systems as complex entities? How well have agent-based modelling approaches lent themselves to simulating the dynamics of spatial processes?

What has been, or will be, the influence of Big Data on increasing our ability to understand and simulate spatial systems? What is the appropriate level of spatial analysis and time frame to model spatial phenomena? We will finish by discussing areas of potential collaboration.


Prof. Susan Stepney (Bio)
Prof. Susan Stepney is Professor of Computer Science, and Director of the York Cross-disciplinary Centre for Systems Analysis YCCSA), at the University of York. She was PI of the EPSRC-funded CoSMoS project, which developed a structured ABM simulation methodology.

The CoSMoS (Complex Systems Modelling and Simulation) approach

I will describe the motivation for our CoSMoS approach, and the details of what it entails. In particular, I will discuss why it is essential to keep the model of the system domain, and the model of the software simulation, carefully separate. I will also give some recent feedback on our use of CoSMoS to develop a simulation of an abstract model of gene expression in the BBSRC-funded project CellBranch.


Dr Paul Richmond (Bio)
Dr Paul Richmond is a Senior Lecturer and Research Software Engineer who has recently been awarded one of only six EPSRC Early Career Research Software Engineering (RSE) Fellowships at the University of Sheffield, UK. The focus of his fellowship is in facilitating the use of accelerated architectures such as Graphics Processing Units (GPUs) to accelerate scientific discovery. He is the lead software developer of the FLAME GPU framework and has a proven
track-record of forming inter-disciplinary collaborations to achieve agenda-driven research. His work focuses on developing software which facilitates the pioneering use of emerging high-performance computing architectures for complex systems simulation within computational science and engineering. He is an excellent communicator with a long term record of engaging scientists, policy makers and the public around the challenges of parallel computing.

From biological cells to populations of individuals: Complex Systems Simulations with CUDA

Complex systems are prevalent throughout various levels of biology from the molecular and cellular scales through to populations of interacting individuals. This talk discusses how formal state based representation of agents within a complex system can be simulated and visualised at large scales using the open source FLAME GPU framework. Methods of code generation from XML documents and use of CUDA streams for heterogeneous state execution are presented. Examples include cellular tissue modelling and large scale crowd dynamics.

Agent Based Modelling (ABM) is a powerful simulation technique used to assess and predict group behaviour from a number of simple interacting rules between communicating autonomous agents. Traditional ABM toolkits are primarily aimed at a single CPU architecture, with an inherent lack of parallelism and a reliance on deeply serialised algorithms which are not well suited to many core architectures. As a result, the uptake of GPUs has had a relatively low impact on the field of ABM. The few examples of ABM on the GPU are limited to specific implementations of discrete space (cell) models or swarm systems.

The Flexible Large-scale Agent Modelling Environment for the GPU (FLAME GPU) addresses the performance and architecture limitations of previous work by presenting a flexible framework approach to ABM on the GPU. Most importantly it addresses the issue of agent heterogeneity through the use of a formal state based agent representation. This representation allows agents to be separated into associated state lists which are processed in batches to allow very diverse population of agents whilst avoiding large divergence in parallel code kernels.

In previous work FLAME GPU has been used in a wide range of research applications from the simulation of cellular level systems biology models of wound formation to pedestrian simulation. The overall aim of FLAME GPU is to provide a technique that allows modellers to harness parallelism and the power of the GPU, without requiring background knowledge of parallelism or the hardware (a domain specific modelling tool). Performance rates equalling, or bettering that of HPC clusters can easily be achieved, with obvious cost to performance benefits. Massive population sizes can be simulated, far exceeding those that can be computed (in reasonable time constraints) within traditional ABM toolkits. Finally the use of the GPU allows simulations to be visualised in real time, allowing interaction and real time steering.

FLAME GPU is currently undergoing a major update which includes the addition of the use of atomics to improve sorting performance for building spatial data structures, CUDA streams to permit simultaneous execution of differing agent populations (or state configurations) and a runtime API to simplify code generation and manage simultaneous simulations (in order to explore sets of parameters). The expected results of the following updates are that agents with diverse states should receive vastly improved performance over the already impressive results compared to existing ABM simulators.

In summary this talk will give an overview of the FLAME GPU approach to state based agent simulation, discuss agent communication algorithms and recent improvements using atomics, present methods for code generation from XML model descriptions and give late breaking performance results for simulations of cellular populations and crowds of pedestrians each with over a million agents.


Dr Kieran Alden (Bio)
Kieran Alden is a Research Fellow at York’s Department of Electronic Engineering and Computational Immunology and Robotics Laboratories. He studied Computer Science and Computational Biology at Royal Holloway and York respectively, and completed his PhD within York’s Centre for Immunology and Infection. He conducts interdisciplinary research that examines and develops engineering processes and tools through which simulations can be developed and analysed, to help us understand real systems and increase confidence in using models to guide experimentation.  Historically this research has focused on modelling immunological and microbiological systems, with more recent focus on engineering models for applications in robotics and economic analyses of healthcare provision.

Using Statistical Approaches and Machine Learning Techniques to Engineer and Understand Agent-Based Simulations

Developing simulations for immunological studies has long been a focus of research in the York Computational Immunology Lab, with the application of agent-based modelling providing relevant tools that have helped interrogate human biology and reduce the reliance on animal experimentation. Although offering significant advantages in capturing individual heterogeneity and incorporating stochasticity, sophisticated agent-based models become more difficult to design, execute, and analyse, making it more challenging to translate predictions into increased understanding. This talk will provide an overview of the simulation work we have conducted, and examine how we are applying and integrating statistical approaches and machine learning techniques with agent-based modelling, to permit enriched model evaluation and refinement and aid better understanding of the design, organisation, dynamics, and function of a modelled system of interest.


Dr Dawn Walker (Bio)
Dr Dawn Walker is a senior lecturer in Computer Science (University of Sheffield) applying computational modelling to biological tissues. She uses agent-based techniques to study cellular interactions in several tissue types and has expanded cellular length scale models to multiscale, multi-paradigm applications including ODE-based models which capture sub-cellular signalling. DCW has collaborated in EU-funded projects on the development of computational frameworks for multiscale-multiparadigm simulation, including  CoAST (Complex Automata Simulation Technique) and CHIC which is specifically focussed patient-specific cancer simulation ( .

Agent-based modelling (ABM) of cellular interactions: three case studies

Interactions between individual biological cells are key to determining health and disease in biological tissues. Furthermore, cellular level behaviour and interaction underlie other outcomes in living systems, such as the successful fertilisation in the human reproductive system.

This talk will focus on three case studies showcasing past and present work in my group, each involving the use of ABMs to understand or simulate emergent biological behaviour: growth and wound healing in epithelial monolayers, a mulltiscale ABM of cancerous mutations in the human colon and finally, simulation of the interaction of sperm in the female reproductive tract.


Be the first to comment on "Talk abstracts"

Leave a comment