ACC C yfronet AGH University of Sheffield Collaboration Marian Bubak Academic Computer Cyfronet Department of Computer Science AGH University of Science and Technology Krak ID: 931865
Download Presentation The PPT/PDF document "From virtual labs to the center of excel..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
From virtual labs to the center of excellence for personalized diagnostics and medical therapyACC Cyfronet AGH – University of Sheffield Collaboration
Marian Bubak Academic Computer Cyfronet, Department of Computer Science, AGH University of Science and Technology, Kraków, PolandRod HoseDepartment of Infection, Immunity and Cardiovascular DiseaseUniversity of Sheffield, United Kingdomhttp://dice.cyfronet.pl/; bubak@agh.edu.pl
Slide2Recent trendsEnhanced scientific discovery is becoming collaborative and analysis-focused; in-silico experiments are becoming more and more complexAvailable compute and data resources are distributed and heterogeneous Modelling of complex collaborative scientific applicationsdomain-oriented semantic descriptions of modules, patterns and data to automate composition of applicationsStudying the dynamics of distributed resources
investigating temporal characteristics, dynamics, and performance variations to run applications with the desired quality of serviceModelling and designing a software layer to access and orchestrate distributed resourcesmechanisms for aggregating multi-format/multi-source data into a single coherent schema semantic integration of compute/data resourcesData-aware mechanisms for resource orchestrationenabling reusability based on provenance dataOur research interests
Slide3Expanding the Grid for a new category of applications in
medicine, environmental control, and physics, running interactively, and extending the Grid infrastructure across eleven European countries. Efficient development of these kinds of applications on the Grid required new tools for verification of source code, performance prediction, evaluation and on-line analysis. The Grid was equipped with new components for monitoring of application performance, efficient distributed data access, specific resource management, as well as portals and mobile personalised user interfaces.
EU
FP5
Project –
CrossGrid
, 2002-2005
http://cordis.europa.eu/pub/ist/docs/grids/crossgrid_achievement.pdf
CrossGrid
:
Grid Environment for Interactive Applications
Slide4Design of a laboratory for virologists, epidemiologists and clinicians investigating the HIV virus and the possibilities of treating HIV-positive patients
Based on notion of in-silico experiments built and refined by cooperating teams of programmers, scientists and clinicians
6FP EU Project –
ViroLab
, 2006-2009
http://www.virolab.org/
Employed full concept-prototype-refinement-production circle for virology tools
Set of dedicated yet interoperable tools bind together programmers and scientists for a single task
Support for system-level science with concept of result reuse between different experiments
ViroLab
Virtual Laboratory
UK partner: UCL London
Slide5MAPPER Memory (MaMe) a semantics-aware persistence store to record metadata about models and scales
Multiscale Application Designer (MAD) visual composition tool transforming high level description into executable experimentGridSpace Experiment Workbench (GridSpace) execution and result management of experiments MADGridSpaceMaMeA method and an environment for composing multiscale applications from single-scale models
Validation of the
the
method
against
real
applications structured using tools
Extension of application composition techniques to multiscale simulationsSupport for multisite execution of multiscale simulationsProof-of-concept transformation of high-level formal descriptions into actual execution using e-infrastructuresUK partner: UCL LondonMAPPER: Multiscale Programming and Execution
7
FP EU Project –
MAPPER
, 2
010-2013
http://www.
mapper-project.eu
Slide6Facilitate
creation, deployment and robust operation of Early Warning Systems in virtualized cloud environmentCommon Information Spaceconnects distributed component into EWS and deploy it on cloudoptimizes resource usage
taking into
acount
EWS
importance
level
provides
EWS and self-monitoringequipped with autohealing
UK partner: HR WallingfordUrbanFlood: Common Information SpaceEarly Warning System (EWS): any system which operates according to four steps: monitoring, analysis, judgment, action (e.g. environmental monitoring)
7
FP EU Project –
UrbanFlood
, 20
10
-20
12
http://www.
urbanflood.eu/
Slide7VPH-Share: Virtual Physiological Human: Sharing for Healthcare – a Research Environment
VPH-Share has developed the organisational fabric (the infostructure) and integrated optimised services to expose and share data and knowledge, jointly develop multiscale models for the composition of new VPH workflows and facilitate collaboration within the VPH communityVPH-Share provides essential services, as well as the computational infrastructure, for the sharing of clinical and research data and tools, facilitating the construction and operation of new VPH workflows, and collaborations between the members of the VPH community. UK partners: University of Sheffield, Sheffield Teaching Hospital, King’s College London, UCL London, NHS Health&Social Care IC, The Open University
7
FP EU Project –
VPH-
Share
, 20
11-2015
http://www.
vph-share.eu/
Slide8VPH
Infostructure
Slide9VPH-Share federated cloud
VPH-Share services hostportal.vph-share.euCore Services Hostvph.cyfronet.pl
Atmosphere-VPH
Secure RESTful API
(Cloud Facade)
Atmosphere core (internal dependency)
VPH-Share cloud site at CYF
Worker nodes
Template repository
VPH-Share cloud site at UNIVIE
Worker nodes
Template repository
MS Azure VPH-Share
cloud
account
Worker nodes
Template repository
RackSpace VPH-Share
cloud
account
Worker nodes
Template repository
Google Compute VPH-Share
cloud
account
Worker nodes
Template repository
Amazon EC2 VPH-Share site
Worker nodes
Template repository
Amazon EC2 CISTIB
site
Worker nodes
Template repository
Slide10EurValveValvular heart disease currently affects 2.5% of the population; it is overwhelmingly a disease of the elderly. EurValve will implement, test and validate a modelling based decision support system (DSS) for aortic and mitral valve diseases that allows simulating, comparing and understanding the effects and risks of different treatment strategies. The DSS will improve knowledge of disease mechanisms by applying a holistic assessment of cardiovascular function that includes haemodynamic data at all cardiovascular compartments (ventricle, valve, vessels) and multiscale components that couple organ with cell function.
H2020 EU Project – EurValve, 2016-2019 http://www.eurvalve.eu
UK partners: University of Sheffield, Sheffield
Teaching
Hospital
, University of Bristol
Slide11EurValve: Model Execution Environment
API – Application Programming Interface REST – Representational state transferRimrock – servis used to submit jobs to HOC cluster Atmosphere – provides access to cloud resourcesgit – a distributed revision control system
Slide12PRIMAGEPRedictive In-silico Multiscale Analytics to support cancer personalized diaGnosis and prognosis (PRIMAGE) is a newly approved project proposal
in which ACC Cyfronet AGH will collaborate with USFD.The goal of the project is to develop an environment for predictive, personalized medicine, focused on cancer treatment.Data infrastructures, imaging biomarkers and models for in-silico medicine research will be validated during this project in the context of two pediatric cancers: Neuroblastoma (NB, the most frequent solid cancer of early childhood) and Diffuse Intrinsic Pontine Glioma (DIPG, the leading cause of brain tumor-related death in children).ACC Cyfronet AGH will be responsible for delivering the required computational resources and tools, enabling implementation and deployment of the platform.The project is set to commence in December 2018 and will last four years.
H2020
EU Project –
PRIMAGE
, 2
018-2022
UK partner: University of Sheffield
Slide13Centre for New Methodsin Computational Diagnostics and Personalised Therapy
Slide14Mission of CECM ProjectCECM EU H2020 „Teaming for Excellence” project has
developed a Business Plan to establish in Kraków a European Centre of Excellence for computational medicine.The CECM Consortium is going to build a world-class centre of excellence, attractive to foreign partners, with a significant impact at both regional and national scales, providing benefits for the pan-European society.It is a consortium of leading European science and innovation institutions in all domains of the new CoE.ACC Cyfronet AGH has a long record of efficient support for scientists in the computational life science EU and PL research projects. Małopolska and Kraków are well positioned for a key role in the computational medicine.
Slide15Leading European science and innovation institutions:University of Sheffield and Insigneo Institute
– experts in translation of in silico modelling and simulations to clinicsForschungszentrum Jülich – experts in modern HPC and data techniques, applied for science and industryFraunhofer ISI – experts in systemic multi-domain solutions and innovation in healthcareThey will work together with Partners from Poland:ACC Cyfronet AGH – experts in simulation and provisioning computing infrastructure for scienceKlaster LifeScience Kraków – Poland’s top cluster of industry, academia and hospitals for the life science domainNCBiR - the Polish National Centre for Research and Development (project coordinator)CECM Partners
Slide16Complexity
The Challenge
The Solution
In
Silico
Medicine
Complexity by composition
Unlimited Capacity
Subject-specific
Precise DiagnosisRanked TreatmentsIntegrated CareAgeing Co-morbiditiesSpecialists’ CapacityPopulation-specificImprecise DiagnosisSuboptimal TreatmentFragmented CareCECM: the Challenge and the Solution
Slide17Objectives of the new CoEA: Development of new computation-based solutions for diagnostics and therapy in daily healthcare
. B: Systematic involvement of regional biomed businesses, specialising in technologies and services for personalised medicine, in high-profile research projects and clinical adoption of their outcome. C: Development of education initiatives to train knowledge workers with the skills in data analytics, simulation, and HPC/Big Data, to respond to the growing demand for skilled workforce in medical devices and bio-engineering.D: Strong advancement of algorithms, models and technologies involved in personalised medicine, including design of holistic, replicable, generic framework for simulation-based Decision Support Systems (DSS) creation.
Slide18A strategic cooperation between Poland & UK
Strongly supported by expertise from Germany Geographically located in KrakówServing the world Dual-centred, with access to Clinic and HPCEnsuring complete scientific coverage
Staffed by experts in Computational Medicine
Handpicked for their skills, fields, compatibility
Kick-started
with EU and Polish Funding, supporting core activities and research
Building the Polish Decision-Support Industry
Championing global healthcare in the 21
st
Century
Self-sustaining
after the first 7 years
Goal
: Leadership in Clinical Decision Support
Slide19AGH Technical University
36,000 students
Cyfronet (Est 1973)
Polish leader in HPC
University
Hospital
Krak
ó
w’s Teaching Hospital
University of Sheffield
29,000 students
Insigneo (Est 2012)
140 researchers, 50 clinicians
ST Hospital Trust
Sheffield’s Teaching Hospitals
Insigneo
:
largest In Silico Medicine (ISM) research institute in Europe
Researchers develop dozen
s
of ISM prototypes
Lacks the capability to
convert
them into industrial-quality solutions with appropriate level of service
Brexit will make this even harder
Cyfronet
mission is to deliver industrial-quality solutions with appropriate level of service
DICE team has seeded a level of expertise in ISM unusual in most HPC centres, but lacks portfolio
Kraków
partners
offer the skills, infrastructures, and development policies to sustain unlimited growth
The Synergy
Slide20Clinical Data
Healthcare InnovationHardwareHPC skillMedical LabsData and TrialsComputer science researchHPC Lab and hardwareComp Medicine skillsHuman resourcesLifeSci/Med research
Human resources
Human resources
Industry
Clinical Data
Technology
The
Centre
An independent research foundation
Venture
capital
and
seed
funds
€ Private€ PublicScience
Advice
Infrastr
’
Talents
Fund/
Exp’t
Data
Modelling research
In silico
expertise
Core
Consortium
Tech/Clinical
Funding
The Consortium,
t
he Partnership
Slide21Collaboration
: Modellers + Healthcare Professionals identify in silico opportunitiesInnovation: New in silico solutions fordiagnostics and therapy in daily healthcare
Exploitation:
Regional biomed businesses
State-of-the-art:
Advancement of algorithms, models and technologies
Education:
knowledge transfer in health data analytics, modelling, HPC
Foster:
young Entrepreneurs technical/commercial/financial skillsTowards a successful, sustainable CoE
Slide22Healthcare Informatics
DataScienceAlgorithmic decision scienceIn silico techniquesModelling and SimulationComputer Science and HPCTheCentreRecruitment of Life Science and IT research and management staffTrack record (2017):8 international CSOs/lab leaders recruitedInternational committees and management : excellence in talent acquisition
Search Committee:
6 members (UK, DE, PL)
Access to
quality candidates
Centre’s Management :
Led by
Scientific Affairs Director
Marian
BubakFoundation Council5 reps of Teaming partners (UK, DE, PL) Concludes employment contractsInt. Scientific Committee : Chaired by Marco Viceconti13 members (UK, IT, DE, NL, US, PL)Selects top candidatesThe Network
Top international candidates for the Directorship, and Laboratory Leaders
Access to local and international Candidates for Middle and Junior Researcher positions
Key Performance Indicators
KPI-1
Research groups
KPI-4
R&D personnel
KPI-5
New scientists
KPI-6
Foreign scientists
KPI-7
Publications
(5)
(55)
(33)
(15)
(60)
Human
Resources
:
Key
to
Success
Slide23Other profits arising from collaboration with the University of SheffieldComplementary character of our collaboration Sound scientific progressCyfronet has become a
member of the VPH Institute and an associate partner of CompBioMedEfficient sharing of computational resources across bordersPorting results of our collaboration to the Polish community (e.g. Atmosphere in PLGrid and ISMOP)Prof. Hose’s lectures at the Department of Computer Science, AGH Kraków
Slide24Collaboration between Sheffield University and Cyfronet has resulted in a unique opportunity to build a world-class centre of excellence, attractive to foreign partners and capable of achieving significant impact
on both regional and national scales, with lasting benefits for the pan-European society.http://dice.cyfronet.pl bubak@agh.edu.pl Summary