Introduction to FAIR Data and Research Objects

FONDO EUROPEO DE DESARROLLO REGIONAL (FEDER)
Una manera de hacer Europa
Métricas FAIR. Principios FAIR y su aplicación en ASIO
Diego López-de-Ipiña
MORElab research group, Universidad de Deusto
dipina@deusto.es

Temas a tratar
 FAIR Data
o Principios
o FAIR vs Open Data
o Implementing FAIR & FAIRmetrics
o FAIRness de ASIO-HERCULES
 Research Objects
o Definición
o Estándar RO-CRATE
o Ejemplos de uso

What is FAIR data?
 A set of principles to ensure that data are shared in a way that enables & enhances reuse, by
humans and machines

Origins of FAIR Data
 Emerged from a workshop held in Leiden in 2014
 Come from life sciences but intended for all data
o Issued by FORCE11 community
 Echo previous principles on Open Research Data
o Data underpinning scientific research results that has no restrictions on its access, enabling anyone to
access it.
o Open Science encompasses a variety of practices, usually including areas like open access to publications,
open research data, open source software/tools, open workflows, citizen science, open educational
resources, and alternative methods for research evaluation including open peer review
Beyond Open Access to publication: https://blogs.uef.fi/ueflibrary-
bors/different-ways-to-make-your-research-more-open/

Origins of FAIR Data
 Emerged from a workshop held in Leiden in 2014
 Come from life sciences but intended for all data
o Issued by FORCE11 community
 Echo previous principles on Open Research Data
o Data underpinning scientific research results that has no restrictions on its access, enabling anyone to
access it.
o Open Science encompasses a variety of practices, usually including areas like open access to publications,
open research data, open source software/tools, open workflows, citizen science, open educational
resources, and alternative methods for research evaluation including open peer review

What FAIR Data means?
Findable
F1. (meta)data are assigned a globally unique and eternally
persistent identifier.
F2. data are described with rich metadata.
F3. (meta)data are registered or indexed in a searchable
resource.
F4. metadata specify the data identifier.
Accessible
A1 (meta)data are retrievable by their identifier using a
standardized communications protocol.
A1.1 the protocol is open, free, and universally
implementable.
A1.2 the protocol allows for an authentication and
authorization procedure, where necessary.
A2 metadata are accessible, even when the data are no
longer available.
Interoperable
I1. (meta)data use a formal, accessible, shared, and broadly
applicable language for knowledge representation.
I2. (meta)data use vocabularies that follow FAIR principles.
I3. (meta)data include qualified references to other
(meta)data.
Reusable
R1. meta(data) have a plurality of accurate and relevant
attributes.
R1.1. (meta)data are released with a clear and accessible
data usage license.
R1.2. (meta)data are associated with their provenance.
R1.3. (meta)data meet domain-relevant community
standards.

FAIR Data Checklist
7
 Findable
o Persistent ID
o Metadata online
 Accessible
o Data online
o Restrictions where needed
 Interoperable
o Use standards, controlled vocabs
o Common (open) formats
 Reusable
o Rich documentation
o Clear usage licence

FAIR vs. Open Data
8
 FAIR data does not have to be open
 Data can be shared under restrictions & still be FAIR
 Making data FAIR ensures it can be found, understood and reused
 Open data is a subset of all the data shared

Implementing FAIR
9
 The principles do not specify particular technologies or implementations e.g. semantic web
 FAIR is not a standard to be followed or strict criteria – it’s a spectrum / continuum
 Supporting FAIR will require investment in infrastructure, coordination across initiatives and
engagement with research communities

Implementing FAIR: FAIRsharing
10
 FAIRsharing is a web-based, searchable portal of three interlinked registries, containing both in-house
and crowdsourced manually curated descriptions of standards, databases and data policies, combined
with an integrated view across all three types of resource.
o By registering your resource on FAIRsharing, you not only gain credit for your work, but you increase its visibility outside of
your direct domain, so reducing the potential for unnecessary reinvention and proliferation of standards and databases.
https://fairsharing.org/educational/

Implementing FAIR: FAIR metrics
11
 Implement the FAIR principles into a data assessment tool so that every dataset which is deposited or
reused from any digital repository can be assessed in terms of a score on the principles Findable,
Accessible, Interoperable, and Reusable, using a ‘FAIRness’ scale from 1 to 5 stars.
 Documentation: https://www.slideshare.net/OSFair/osfair2017-workshop-outcomes-fair-metrics-
starring-your-data-sets
 “A design framework and exemplar metrics for FAIRness”,
https://dx.doi.org/10.1038%2Fsdata.2018.118

Comprobación del nivel FAIR en ASIO
12
 Sistema de comprobación automatizado del nivel FAIR cumplido por los recursos publicados durante el
desarrollo del proyecto, que devuelve el nivel FAIR alcanzado, asociado a la versión del proyecto y la
fecha en la que se ejecutó la comprobación.
 La red de ontologías ROH es FAIR significa que:
1. Ofrece datos que sean Findable (Encontrables) a través de un identificador persistente e incluyendo
metadatos
2. Accessible (Accesible) a través del protocolo universal HTTP
3. Interoperable usando vocabularios ampliamente adoptados y
4. Reusable, se publican usando licencias de uso que promocionen la reusabilidad, como por ejemplo Creative
Commons 4.0 BY-SA.
 La mejor manera de publicar datos FAIR es hacerlo mediante Linked Data, teniendo especial cuidado de
generar datos y metadatos de alta calidad, mejorando así la reusabilidad de los datos para máquinas (y,
como consecuencia y en última instancia, para humanos).

Sistema de Comprobación FAIRness Automatizado en HERCULES-ASIO
13
 La definición precisa de las métricas se ha realizado en base a las métricas definidas en el proyecto FAIR
Metrics
 14 métricas formuladas por FAIRmetrics
o https://github.com/FAIRMetrics
 Front-end (interfaz de usuario) que facilite el lanzamiento del test suite y la generación de reportes que
informen sobre el grado de cumplimiento de las métricas FAIR
o http://herc-as-front-desa.atica.um.es/collections/15/evaluate/template
 Tales reportes son archivados (históricos) para poder analizar el compromiso a lo largo del tiempo de
ROH con los principios FAIR.

Estrategia de implementación de métricas FAIR
1. Evaluación del estado de cumplimiento actual de las propiedades FAIR del prototipo ROH:
o Servicio de ejecución de evaluación de servicios FAIR: https://fairsharing.github.io/FAIR-
Evaluator-FrontEnd/#!/collections/new/evaluate
o Resultado: https://fairsharing.github.io/FAIR-Evaluator-FrontEnd/#!/evaluations/4065
2. Análisis de la implementación disponible de las métricas en Ruby
o Cambios introducidos en los últimos meses lo que implica que el proyecto sigue activo:
https://github.com/FAIRMetrics/Metrics/tree/master/MetricsEvaluatorCode/Ruby/metrictests
o Despliegue del back-end de métricas FAIR en máquina local para uso dentro del proyecto:
 GNOSS ha realizado despliegue en el servidor de desarrollo
 Se ha contactado con autores, concretamente mark.wilkinson@upm.es
 Su recomendación ha sido que hagamos uso de la API disponible en https://ejp-
evaluator.appspot.com/FAIR_Evaluator//
 Sin embargo, es necesario tener nuestro propio despliegue para evitar
dependencias de terceros

3. Diálogo con autores de FAIRsharing (Mark Wilkinson):
o Es razonable contribuir al proyecto existente en vez de reinventar la rueda
o Se ha contrastado que las métricas FAIRsharing funcionan. Análisis realizado con:
o Simple recurso apuntando a un DOI (10.1016/j.future.2019.09.033):
https://fairsharing.github.io/FAIR-Evaluator-FrontEnd/#!/evaluations/3236
o Tal como se ha mencionado ejecutado también contra la definición de la red de ontologías
ROH: https://fairsharing.github.io/FAIR-Evaluator-FrontEnd/#!/evaluations/4065
o Es posible crear nuevas métricas específicas al dominio de gestión de datos de la investigación a
través de la compleción de FAIR Maturity Indicator template. A partir de esta plantilla:
o Equipo de FAIRsharing se ofrece a implementarlo
o Se pueden realizar extensiones a través de la provisión de un endpoint para tales nuevas
métricas usando SmartAPI (https://smart-api.info/), extensión de OpenAPI

4. Decisión de estrategia de implementación de métricas FAIR
o Creación de un proxy API usando OpenAPI, que será utilizada desde back-end HERCULES-ASIO
para realizar validación de recursos (resources) y de la red de ontologías como tal
o Tal API se comunico en primera instancia con API pública ya desplegada por FAIRsharing:
https://ejp-evaluator.appspot.com/FAIR_Evaluator// y ofrece interfaz web Swagger
o Incluye interfaz gráfica para introducir, basada en https://github.com/FAIRsharing/FAIR-
Evaluator-FrontEnd:
o Recurso a validar
o Conjunto de métricas a ejecutar
o Visualización amigable del grado de cumplimiento de las políticas FAIR
o Publicación de issues y pull requests con modificaciones a las métricas ya implementadas en el
caso de encontrarse bugs o realizarse extensiones a las existentes
o Clonar API pública de FAIRsharing para asegurar no dependencia con recurso externo
o Generación de nuevas métricas específicas al ámbito ASIO a través de smartAPI y/o entrega de
plantillas para FAIRsharing

5. Solución desarrollada
o Se ha creado un proxy API (bridge) usando OpenAPI, que será utilizada desde HERCULES-ASIO
para realizar validación de recursos (resources) y de la red de ontologías como tal
o Incluye interfaz Swagger con 4 operaciones:
o Recuperar colecciones de métricas (baterías de test disponibles)
o Ejecución de colección de métricas seleccionada sobre recurso indicado
o Recuperación de evaluaciones realizadas desde un ORCID
o Recuperación de resultados de evaluación de ejecución de colección de métricas
6. Trabajo futuro
o Generación de recomendaciones para hacer ROH y recursos asociados (y por ende otras
ontologías y recursos) compatible con políticas FAIR:
https://github.com/FAIRMetrics/Your_Path_To_FAIRness/tree/master/MI_Test_Tutorials
o Ejecución de políticas FAIR sobre test suite de recursos ROH y la red de ontologías
o Adaptación de ROH para su cumplimiento integro de políticas FAIR

Arquitectura puente entre ASIO y FAIRmetrics
ASIO-
FAIRmetrics
Bridge
BBDD
Evaluaciones
ROH
ASIO-HERCULES
Web front-end
FAIRmetrics
back-end
ASIO-HERCULES
back-end
FAIRmetrics
back-end

Bridge ASIO-FAIR
Metrics código
fuente
https://github.com/Hercules
CRUE/GnossDeustoBackend/t
ree/master/src/fair/bridge
http://herc-as-front-
desa.atica.um.es/bridgeswag
ger/v1/ui

Pasos para la ejecución de FAIRness de recursos
1. Identificar y seleccionar batería de tests a realizar
o Se puede usar una de las baterías existentes: https://fairsharing.github.io/FAIR-Evaluator-
FrontEnd/#!/collections
o O definirse una nueva a través de interfaz: https://fairsharing.github.io/FAIR-Evaluator-
FrontEnd/#!/collection/new
2. Instalación y lanzamiento del puente ASIO-FAIRmetrics
o Configuración para selección endpoint FAIRmetrics con el que conectarse
o Acceso al front-end Swagger
3. Ejecución de operaciones sobre API bridge ASIO-FAIRmetrics
o Ejecución de los métodos del API con recursos externos, ej. 10.1109/ACCESS.2019.2952321
o O con la propia red de ontologías de Hércules: http://purl.org/roh

Pasos para la ejecución de FAIRness de recursos (paso 1)

Ejecución de puente Bridge ASIO-FAIR Metrics (paso 2)

Ejecución de
colección de
métricas en
puente bridge
ASIO-FAIR
Metrics (paso 3)
http://herc-as-front-
desa.atica.um.es/bridg
eswagger/v1/ui/#/defa
ult/collections_id_evalu
ate_post

Validación FAIRness publicación con puente ASIO-FAIRmetrics

Validación FAIRness Red de Ontologías Hercules (ROH)

Research Objects
 An emerging approach to the publication, and exchange of scholarly information on the Web.
 Research Objects aim to improve reuse and reproducibility by:
o Supporting the publication of more than just PDFs, making data, code, and other resources first
class citizens of scholarship
o Recognizing that there is often a need to publish collections of these resources together as one
shareable, cite-able resource.
o Enriching these resources and collections with any and all additional information required to
make research reusable, and reproducible!
Science advances on a foundation of trusted discoveries. Reproducing an experiment is one important
approach that scientists use to gain confidence in their conclusions.
Marcia McNutt, Editor-in-Chief of Science

Research Objects: going beyond PDFs
 The goal of research objects is to improve the potential
for understanding and reuse of research by making sure
that the information that is needed to make a published
resource useful is associated with it and shared as a
whole.
 As part of one research investigation, you might for
example have:
o Slides hosted on slideshare,
o Code in a GitHub repository,
o Data in figshare,
o Data in ArrayExpress.
o A research object not just a publication but everything
else that goes into, and supports an investigation
o Initiatives and approaches to bring content together in a
machine-readable mechanism so that it can be more
easily shared and exchanged.

Research Objects: principles
 Identity:
o Use globally unique identifiers as names for things. For example DOI’s for publications or ORCID IDs
for researchers. This is for two reasons
1. So that we can talk unambiguously about things.
2. So that people can find those things.
 Agregation
o Use some mechanism of aggregation to associate things that are related or part of the broader
investigation, study, etc.
• Core to the value of Research Objects - providing the supporting artefacts that make the research
potentially useful to someone else.
 Annotation:
o Provide additional metadata about those things, how they relate to each other, where they came
from, when etc.
• This helps people discover data that are relevant and potentially useful.

Research Objects: benefits
 A way to refer to (and cite!) a bundle of research artefacts as a whole.
 A mechanism to associate human and machine-readable metadata with those research
artefacts. Anything that can understand that metadata can therefore benefit.
 A graceful degradation of understanding.

Research Objects: modelling ROs with RO-CRATE
 RO-Crate is a community effort to establish a lightweight approach to packaging
research data with their metadata.
o It is based on schema.org annotations in JSON-LD.
 The RO-Crate effort brings together practitioners from very different backgrounds, and
with different motivations and use-cases. Among our core target users are:
1. researchers engaged with computation and data-intensive, workflow-driven analysis;
2. digital repository managers and infrastructure providers;
3. individual researchers looking for a straight-forward tool or how-to guide to “FAIRify” their data;
4. data stewards supporting research projects in creating and curating datasets.
 RO-CRATE metadata specification: https://www.researchobject.org/ro-crate/1.1/
o Reference: Eoghan Ó Carragáin; Carole Goble; Peter Sefton; Stian Soiland-Reyes (2019): A
lightweight approach to research object data packaging Bioinformatics Open Source Conference
(BOSC2019) https://doi.org/10.5281/zenodo.3250687

Research Objects: tools to ROs with RO-CRATE
 Describo interactive desktop application to create, update and export RO-Crates for
different profile
 More tools at: https://www.researchobject.org/ro-crate/implementations.html
o Adopters of RO-Crate are free to supplement RO-Crate using Schema.org metadata and/or assertions using
other Linked Data vocabularies.

Research Objects: RO structure
 The structure an RO-Crate MUST follow is:
<RO-Crate root directory>/
| ro-crate-metadata.json # RO-Crate Metadata File MUST be present
| ro-crate-preview.html # RO-Crate Website homepage MAY be present
| ro-crate-preview_files/ # MAY be present
| | [other RO-Crate Website files]
| [payload files and directories] # 0 or more
 Documented in https://www.researchobject.org/ro-crate/1.1/metadata.html

Research Objects: RO metadata
 Documented in https://www.researchobject.org/ro-crate/1.1/metadata.html
 The Root Data Entity is a Dataset that represent the RO-Crate as a whole; a Research
Object that includes the Data Entities and the related Contextual Entities.
 The primary purpose for RO-Crate is to gather and describe a set of Data entities in the
form of (linked with the root entity with hasPart property):
o Files
o Directories
o Web resources
 The data entities can be further described by referencing contextual entities such as
persons, organizations and publications.
o Contextual entities however primarily exist outside the digital sphere (e.g. People,
Places) or are conceptual descriptions that primarily exists as metadata, like
GeoCoordinates and ContactPoint.

Research Objects: example 1
{ "@context": "https://w3id.org/ro/crate/1.1/context",
"@graph": [
{
"@type": "CreativeWork",
"@id": "ro-crate-metadata.json",
"conformsTo": {"@id": "https://w3id.org/ro/crate/1.1"},
"about": {"@id": "./"}
},
{
"@id": "./",
"@type": [
"Dataset"
],
"hasPart": [
{
"@id": "cp7glop.ai"
},
{
"@id": "lots_of_little_files/"
}
]
},
{
"@id": "cp7glop.ai",
"@type": "File",
"name": "Diagram showing trend to increase",
"contentSize": "383766",
"description": "Illustrator file for Glop Pot",
"encodingFormat": "application/pdf"
},
{
"@id": "lots_of_little_files/",
"@type": "Dataset",
"name": "Too many files",
"description": "This directory contains many small files, that we're not going to describe in detail."
}
]
}
<RO-Crate root>/
| ro-crate-metadata.json
| cp7glop.ai
| lots_of_little_files/
| | file1
| | file2
| | ...
| | file54

Research Objects: example 2
{ "@context": "https://w3id.org/ro/crate/1.1/context",
"@graph": [
{
"@id": "ro-crate-metadata.json",
"conformsTo": {"@id": "https://w3id.org/ro/crate/1.1"},
"about": {"@id": "./"}
},
{
"@id": "./",
"identifier": "https://doi.org/10.4225/59/59672c09f4a4b",
"@type": "Dataset",
"datePublished": "2017",
"name": "Data files associated with the manuscript:Effects of facilitated family case conferencing for ...",
"description": "Palliative care planning for nursing home residents with advanced dementia ...",
"license": {"@id": "https://creativecommons.org/licenses/by-nc-sa/3.0/au/"}
},
{
"@id": "https://creativecommons.org/licenses/by-nc-sa/3.0/au/",
"description": "This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Australia License. To
view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/3.0/au/ or send a letter to Creative Commons, PO Box
1866, Mountain View, CA 94042, USA.",
"identifier": "https://creativecommons.org/licenses/by-nc-sa/3.0/au/",
"name": "Attribution-NonCommercial-ShareAlike 3.0 Australia (CC BY-NC-SA 3.0 AU)"
}
]
}
Documentación de contextual entities:
• https://www.researchobject.org/ro-crate/1.1/contextual-
entities.html

Red Ontologías Hércules – ROH
Diego López-de-Ipiña
MORElab research group, Universidad de Deusto
dipina@deusto.es

Introduction to FAIR Data and Research Objects

More Related Content

What's hot (20)

Similar to Introduction to FAIR Data and Research Objects (20)

More from Diego López-de-Ipiña González-de-Artaza (20)

Recently uploaded (20)

Introduction to FAIR Data and Research Objects