Workshop agenda

Day 1: Tuesday 6th November 2018
An introduction to DataSHIELD

Time Description
09:00-09:30 Registration and refreshments
09:30-09:45 Talk: Welcome address
Prof Paul Burton, Newcastle University
09:45-10:45 Keynote Talk: TBC
Prof Madeleine Murtagh, Newcastle University
10:45-11:15 Morning Break
11:15-12:30 Talk and discussion: Introduction to DataSHIELD
Prof Paul Burton, Newcastle University
12:30-13:30 Lunch
13:30-14:15 Software Tutorial*: Introduction to R
Becca Wilson and Demetris Avraam, Newcastle University
* Please bring a laptop to participate in this tutorial
14:15-15:15 Software Tutorial*: DataSHIELD User Experience
Becca Wilson and Demetris Avraam, Newcastle University
* Please bring a laptop to participate in this tutorial
15:15-15:45 Afternoon Break
15:45-16:45

Talks: DataSHIELD usecases

DataSHIELD-Interconnect usecase
Tom Bishop, MRC Epidemiology Unit, University of Cambridge 

Preliminary results of a federated analysis of cross-sectional data from ENPADASI studies: associations of protein and fat intake with HDL-cholesterol in subjects aged 13 to 80 years
Mariona Pinart, Max Delbrück Center for Molecular Medicine in the Helmholtz Association
16:45-17:00 Wrap up and close Day 1

Day 2: Wednesday 7th November 2018
DataSHIELD infrastructure

Time Description
09:00-09:30 Registration and refreshments
09:30-10:30 Keynote talk: Connected Health Cities and Great North Care Record
Prof Paul Burton, Newcastle University
10:30-10:45 Talk: DataSHIELD usecase - Connected Health Cities North East
Olly Butters, Newcastle University
10:45-11:15 Morning Break
11:15-12:30 Discussion Session: Management of DataSHIELD infrastructure at scale
Chair: Becca Wilson, Newcastle University 
Contributions from: Stuart Wheater and Olly Butters, Newcastle University

Discussion points:
  • provisioning of DataSHIELD
  • DataSHIELD CRAN update
  • user management
  • documentation
  • monitoring
  • standardised procedures for setting up datasets
  • identify pressure points
12:30-13:30 Lunch
13:30-15:15

Infrastructure Talks and discussion:

A queue-Poll Extension - standardised, monitored, indirect and secure DataSHIELD access to your data
Julian Gründner, University of Erlangen-Nuremberg

Deep Learning with Boltzmann Machines in DataSHIELD
Stefan Lenz, University of Freiburg

Discussion Session: DataSHIELD bottlenecks - working with big data
Chair: TBC

Discussion points:

  • identify pressure points
  • restrictions in R/compute specifications
  • working with 'omics data
  • machine learning 
15:15-15:45 Afternoon Break
15:45-16:45 Discussion Session: DataSHIELD integration
Chair: Paul Burton, Newcastle University
Contributions: Yannick Marcon, Opal

Discussion points:
  • Current and future Opal developments
  • generalisation of DataSHIELD
  • integration with other data management platforms i2b2, TransMART, MOLGENIS
16:45-17:00 Wrap up and close Day 2
19:00 Workshop dinner 

Day 3: Thursday 8th November 2018
New methodology and functionality

Time Description
09:00-09:30 Registration and refreshments
09:30-10:30 Discussion session: Contributing to DataSHIELD
Chair: TBC
Contributions: Becca Wilson, Olly Butters, Stuart Wheater, Demetris Avraam, Prof Paul Burton, Newcastle University
  • DataSHIELD governance
  • How to contribute
  • Code testing and auditing
  • Giving credit
10:30-11:00 Morning break
11:00-12:00 Talk: New user functionality
Demetris Avraam, Newcastle University
12:00-12:30

Talk: New developer functionality
Prof Paul Burton, Newcastle University
12:30-13:30 Lunch
13:30-15:00

Talks: New statistical methods

Recovery of IPD inferences from key IPD summaries only: Applications within DataSHIELD
Federico Bonofiglio, University of Freiburg

Distributed regression modelling for selecting markers in DataSHIELD
Daniela Zoller, University of Freiburg

Distributed Statistical Analysis Using DataSHIELD in Epidemiological Studies
Rui Camacho, INESC TEC, University of Porto
15:00-15:30 Afternoon Break
15:30-16:30 Discussion session: What functionality is missing from DataSHIELD
  • what functions do users need from DataSHIELD
  • source functions from outside R
  • new data types
  • prioritise new developments
16:30-17:00 Wrap up and close Day 3