EPSRC eScience Pilot Project in Integrative Biology David Gavaghan, Damian Mac Randal, and Sharon Ll

1
(No Transcript)
2
EPSRC e-Science Pilot Project in
Integrative BiologyDavid Gavaghan, Damian Mac
Randal, and Sharon Lloyd
3
Project Overview

• Focus of first round of UK e-Science Projects
• Data storage, aggregation, and synthesis
• Life Sciences projects focused on supporting the
  data generation work of laboratory-based
  scientists
• Key goal now is to turn this wealth of data into
  information that can be used to determine
  biological function
• Requires an iterative interplay between
  experiment, mathematical modelling, and
  HPC-enabled simulation
• Primary goal of this project is to build the
  necessary Grid infrastructure to support this
  goal

4
The Science and e-Science Challenge

• To build an Integrative Biology Grid to support
  applications scientists addressing the key
  post-genomic aim of determining biological
  function
• To use this Grid to begin to tackle the two
  chosen Grand Challenge problems the in-silico
  modelling of heart failure and of cancer.

5
Two Grand Challenge Research Questions

• What causes heart disease?
• How does a cancer form and grow?
• These two diseases together cause 61 of all
  deaths in the UK

6
Courtesy of Peter Kohl (Physiology, Oxford)
Normal beating
Fibrillation
7
Multiscale modelling of the heart
MRI image of a beating heart
Fibre orientation ensures correct spread of
excitation
Contraction of individual cells
Current flow through ion channels
8
Simulation of sudden cardiac death due to a
mechanically induced impact applied during
repolarisation
Courtesy of W.Li, P.Kohl, and N.Trayanova.
J. Mol. Hist. 2004 (in press)
Required 27 hours of CPU time on an SGI IRIX 64
9
Mathematical model of a beating heart by the
Auckland Group
10
Multiscale modelling of cancer
11
An integrative approach to disease modelling?

• The potential impact of this approach has been
  demonstrated by the work on modelling the heart
• Time is ripe to extend to cancer UK has
  extensive expertise but little has yet been done
• Together the two application areas provide a
  sufficiently hard e-Science problem to require a
  generic solution
• Methodology and infrastructure will be utilised
  across biology and in other scientific domains

12
The scientific challenge

• Modelling and coupling phenomena which occur on
  many different length and time scales

• 1m person
• 1mm tissue morphology
• 1mm cell function
• 1nm pore diameter of a membrane protein
• Range 109

• 109 s (years) human lifetime
• 107 s (months) cancer development
• 106 s (days) protein turnover
• 103 s (hours) digest food
• 1 s heart beat
• 1 ms ion channel gating
• 1 ms Brownian motion
• Range 1015

13
Details of test-run of heart simulation code on
HPCx

• Modelled 2ms of electrophysiological excitation
  of a 5700mm3 volume of tissue from the left
  ventricular free wall
• Noble 98 cell model used
• Mesh contained 20,886 bilinear elements (spatial
  resolution 0.6mm)
• 0.05ms timestep (40 timesteps in total)
• Required 978s CPU on 8 processors and 2.5 Gbytes
  of memory
• A complete simulation of the ventricular
  myocardium would require up to 30 times the
  volume and at least 100 times the duration
• Estimated max compute time to investigate
  arrhythmia 107s (100 days) requiring 100Gb of
  memory (compute time scales to the power 5/3)
• At high efficiency this scales to approximately 1
  day on HPCx

14
Key Deliverables

• A robust and fault-tolerant infrastructure to
  support post-genomic research in integrative
  biology that is user and application driven
• 2nd Generation Grid bringing together components
  across range of current EPSRC pilot projects

15
The e-Science Challenge

• To leverage the global Grid infrastructure to
  build an international collaboratory which
  places the applications scientist within the
  Grid allowing fully integrated and collaborative
  use of
• HPC resources (capacity and capability)
• Computational steering, performance control and
  visualisation
• Storage and data-mining of very large data sets
• Easy incorporation of experimental data
• User- and science-friendly access
• gt Predictive in-silico models to guide
  experiment and, ultimately, design of novel
  drugs and treatment regimes

16
e-Science/Grid Research Issues

• Ability to carry out reliably and resiliently
  large scale distributed coupled HPC simulations
• Ability to co-schedule Grid resources based on a
  GGF-agreed standard
• Use of Grid Services based on OGSA-DAI for data
  virtualisation
• Secure data management and access-control in a
  Grid environment
• Grid services for computational steering
  conforming to an agreed GGF standard

17
e-Science/Grid Research (contd.)

• Grid Services for supporting distributed
  collaborative working including steering and
  visualisation
• An interface to using Grid resources which
  understands and supports effectively the science
  context of the project
• The project also stretches the cross-disciplinary
  aspects of the Grid by linking medical,
  biological, engineering and computing activities.
• The project is intending to produce a long term
  (10 year) production environment based on the
  Grid to support what we expect to become a major
  scientific growth area.

18
Architecture and Software Engineering

• Initially use Web Services to provide a platform
  and language independent interface to the main
  functional components
• Adopt Grid Services as stable open source
  OGSA-compliant implementations become available
• Deploy an object-oriented component-based toolkit
  allowing a plug-and-play style programming
  paradigm
• Use of Portal Technologies to provide
  collaborative access to services

19
Architecture
20
Architecture
21
Technology Gaps that will be addressed

• Much of this work will be in conjunction with
  other EPSRC Pilot projects
• Resilient, robust, reliable Grid framework for
  large scale distributed coupled simulations
• Standardised Grid framework for computational
  steering and visualisation
• Metadata schemas for describing the information
  and data resources involved
• Standardised means to schedule multiple resources
  on the Grid concurrently
• Tools for collaborative working in a Grid
  Services environment
• Transparent Grid

22
Project management

• Building on extensive experience in other
  e-Science projects (particularly e-DiaMoND)
• Focus on team building and common goals (key for
  large, inter-institutional development projects)
• Establishing good communication mechanisms
• Iterative prototype development

23
The Team

• World-leading expertise in the two application
  areas
• IBM
• CCLRC
• Seven UK and NZ Universities (Oxford, Nottingham,
  Leeds, UCL, Birmingham, Sheffield and Auckland)
• Expertise from across the UK e-Science Programme
• Extensive existing connectivity between all
  members of the consortium and with the wider
  research communities in e-Science and within the
  application areas
• Research training in an area crucial to the UK

24
The Resources

• 2.44M from EPSRC e-Science to fund 10 PDRAs and
  6 PhD students
• A further 4 PhD students plus sys admin and
  secretarial support funded internally
• Equivalent of 3FTEs from IBM plus substantial
  hardware discounts to provide a Power 4 server
  and high performance workstations to all project
  staff.
• Use of Atlas Data store at RAL and substantial
  commitment of staff time by CCLRC
• Large pool of expertise through the
  co-investigators in the seven partner
  universities, IBM and CCLRC
• Extensive access to national HPC resources (HPCx
  and CSAR)

25
Current Status

• Award letter issued 26/9/03, agreed by University
  in late October, grant announced 26/10/03.
• Project manager, project architect, six PDRAs,
  and one D.Phil student already appointed
• Project Structure defined and agreed,
  requirements gathering and security policy
  exercises commenced
• Recruitment of other staff in process
• Kick off meeting of project participants held in
  Oxford on January 19th

26
(No Transcript)