Building Capability for Facilities Supported Structural Science

Brian Matthews

Scientific Information GroupE-Science Centre

STFC Rutherford Appleton Laboratory

brian.matthews@stfc.ac.uk

Facilities Support

Big Facilities for Small Science

DLS

ISISCLF

The Science we do - Structure of materials

• ~30,000 user visitors each year in Europe: – physics, chemistry, biology,

medicine, – energy, environmental,

materials, culture– pharmaceuticals,

petrochemicals, microelectronics

Fitting experimental data to model

Bioactive glass for bone growth

Structure of cholesterol in crude oil

Hydrogen storage for zero emission vehicles

Magnetic moments in electronic storageLongitudinal strain in

aircraft wing

Diffraction pattern from sample

Visit facility on research campus

Place sample in beam

• Billions of € of investment– c. £400M for DLS– + running costs

• Over 5.000 high impact publications per year in Europe

– But so far no integrated data repositories

– Lacking sustainability & traceability

I2S2 - Infrastructure for Integration in Structural Sciences

Bridging the gap between raw and derived data

“Lone” researcher scenario• data sharing with colleagues via email• Little or no infrastructure• Little management of raw or derived data

EPSRC National Crystallography Service

• service provision function

• operates across institutions

• moderate infrastructure

Diamond & ISIS•operates on behalf of multiple institutions •processes for experiments •large infrastructure engineered to manage raw data•derived data taken off site on laptops / removable drives

Facilities Lifecycle

Proposal

Approval

SchedulingExperiment

Data cleansing

Record Publication

Scientist submits

application for beamtime

Facility committee approves

application

Facility registers, trains, and schedules

scientist’s visit

Scientists visits, facility run’s experiment

Subsequent publication

registered with facility

Raw data filtered and cleansed

Data analysis

Tools for processing made

available

Metadata Repository

Investigation

Publication KeywordTopic

Sample Sample ParameterDataset

Dataset ParameterDatafile

Datafile Parameter

Investigator

Related Datafile

Parameter

Authorisation

Core Scientific Metadata Model (CSMD)A common general format for Scientific Studies and data holdings metadata

• Cataloguing data holdings• Related to the experiment• Provide access for the Data Owner

• Ease citation, sharing collaboration, and integration

• Allow easy Federation of distributed metadata into a homogeneous Platform

The Core Metadata model forms the information model for ICAT.

Within a Data Policy

Interactions between research process

Grant Proposal

Facilities Proposal

FacilitiesExperiment Data

cleansing

Record Publication

Data analysisLocal

experiments

Simulation

Sample Preparation

Literature Review

Publication

Proposal

Approval

Scheduling

Facilities Experiment

Data storage

Record Publication

Analysis Tools

ORE-CHEM• An abstract model for planning and

enacting chemistry experiments

Earth Sciences: typical workflow

Martin Dove & Erica Yang

• Processing dependent on specialised software

• Sustainability issues • Context not routinely

captured• Main analysis is reliant on

scientist’s knowledge and experience

• selecting parameters and interpreting data

• recorded in a lab note book • Actual workflow not

recorded• Distributed Data - Little

shared infrastructure• Raw and reduced data stored at

ISIS• Other data on his/her laptop or

WebDAV

Interoperability with PublishersIUCr journal policy - “data” either

• must be supplied in CIF format as an integral part of article submission and are freely available for download or

• must be deposited with the Protein Data Bank before or in concert with article publication; the article will link to the PDB deposition using the PDB reference code

Thanks to Brian MacMahon, IUCr

Publication flow in IUCr journals

Experiment

Structure solution

IUCr journals

Chemistry databases

Data reduction

Validation Peer review

editing Publish

Bibliographic databases

RAW

Research Activity Model

A notion of a research activity – a step in the lifecycle model- Can define different types of activity.

Capabilities• Good established formats

– Raw data – e.g. NeXus– Analysed data – e.g. CIF

• Well supported processes for data collection especially at facilities– ICAT and similar tools as unifying medium– Simple metadata models for experiments

• Areas needing work– Upstream planning and synthesis– Downstream analysed data– Sharing and integration

• Drivers– Drive from Facilities (large and small)– Drive from Publishers

Thank You

Questions?brian.matthews@stfc.ac.uk

www.e-science.stfc.ac.uk/