Introduction to FAIR principles in the context of computational biology models. Presented at a Workshop at the Basel Conference of Computational Biology. Grants: European Commission: EOSCsecretariat.eu - EOSCsecretariat.eu (831644)

1. When is a model FAIR –
and why should we care?
Dagmar Waltemath
Basel Computational Biology Conference, Sep 13 2021 | https://www.bc2.ch/
CC BY-NC-ND 3.0
Department of Medical Informatics
University Medicine Greifswald (Germany)

2. © Copyright Universitätsmedizin Greifswald
2
• Identifiable data
items
• Persistent
• Searchable
• Identifiers use standard
protocols
• Authentification
• Access to meta data,
even if data not
accessible
• Formal,
accesssible data
representation
• Qualified
references
• Licensing
• Provenance
• Standards
compliance
FAIR pripnciples published by Wilkinson et al (2016): https://doi.org/10.1038/sdata.2016.18
The FAIR principles in bio* sciences

3. © Copyright Universitätsmedizin Greifswald
FAIR pripnciples published by Wilkinson et al (2016): https://doi.org/10.1038/sdata.2016.18
Finding computational models
3
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource

4. © Copyright Universitätsmedizin Greifswald
BioModels COVID-19 collection: https://www.ebi.ac.uk/biomodels/search?offset=20&numResults=10&sort=relevance-desc&query=COVID-19&domain=biomodels
Finding computational models
4
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource
Example: Globally unique and persistent
model ID in BioModels

5. © Copyright Universitätsmedizin Greifswald
PMR2: Metadata for an exposure https://models.physiomeproject.org/exposure/e5cfb42225d4534a1e08979e57cf8bdd/cloutier_2009_a.cellml/cmeta
Finding computational models
5
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource
Example: Metadata about models in PMR2

6. © Copyright Universitätsmedizin Greifswald
PMR2: Metadata for an exposure https://models.physiomeproject.org/exposure/e5cfb42225d4534a1e08979e57cf8bdd/cloutier_2009_a.cellml/cmeta
Finding computational models
6
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource
Example: Metadata about models in PMR2

7. © Copyright Universitätsmedizin Greifswald
PMR2: Metadata for an exposure https://models.physiomeproject.org/exposure/e5cfb42225d4534a1e08979e57cf8bdd/cloutier_2009_a.cellml/cmeta
Finding computational models
7
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource
Example: Metadata about models in PMR2

8. © Copyright Universitätsmedizin Greifswald
Identifiers.org as a resolution services for URIs in Computational Biology: http://identifiers.org/
Finding computational models
8
Data should be uniquely & persistently
identifiable; researchers should find your data.
F1. (Meta)data are assigned a globally
unique and persistent identifier
F2. Data are described with rich metadata
F3. Metadata clearly and explicitly include
the identifier of the data they describe
F4. (Meta)data are registered or indexed in
a searchable resource
Example: Model repositories and metadata indexed
at identifiers.org

9. © Copyright Universitätsmedizin Greifswald
Identifiers.org as a resolution services for URIs in Computational Biology: http://identifiers.org/
Accessing computational models
9
Conditions under which the data can be used
should be clear (to machines & humans).
A1. (Meta)data are retrievable by their
identifier using a standardised
communications protocol
A2. Metadata are accessible, even when
the data are no longer available
(and experiments)

10. © Copyright Universitätsmedizin Greifswald
Identifiers.org as a resolution services for URIs in Computational Biology: http://identifiers.org/
Accessing computational models
10
Conditions under which the data can be used
should be clear (to machines & humans).
A1. (Meta)data are retrievable by their
identifier using a standardised
communications protocol
A2. Metadata are accessible, even when
the data are no longer available
(and experiments)
Example: Retrieving COVID-19 models from
BioModels
HTTPS SPARQL

11. © Copyright Universitätsmedizin Greifswald
11
Interoperable models across systems
Machine-readable and using terminologies,
vocabularies or ontologies that are commonly
used in the field
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

12. © Copyright Universitätsmedizin Greifswald
SBML L3 V1 Core Annotation Scheme, taken from https://resolver.caltech.edu/CaltechAUTHORS:20130108-162112228 12
Interoperable models across systems
Machine-readable and using terminologies,
vocabularies or ontologies that are commonly
used in the field
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
Example: Annotation of models (archives) using
bioontologies, RDF & following the metadata
specification.

13. © Copyright Universitätsmedizin Greifswald
SBML L3 V1 Core Annotation Scheme, taken from https://resolver.caltech.edu/CaltechAUTHORS:20130108-162112228 13
Interoperable models across systems
Machine-readable and using terminologies,
vocabularies or ontologies that are commonly
used in the field
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
Example: Annotation of models (archives) using
bioontologies, RDF & following the metadata
specification.

14. © Copyright Universitätsmedizin Greifswald
OMEX standard: https://doi.org/10.1515/jib-2020-0020; Harmonised annotations: https://doi.org/10.1093/bib/bby087 14
Interoperable models across systems
Machine-readable and using terminologies,
vocabularies or ontologies that are commonly
used in the field
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
Example: Annotation of models (archives) using
bioontologies, RDF & following the metadata
specification.

15. © Copyright Universitätsmedizin Greifswald
15
Reusing other people‘s models
Well-described with metadata & provenance
information; data sources can be linked or
integrated with other data sources.
R1. (Meta)data are richly described with 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
detailed provenance
R1.3. (Meta)data meet domain-relevant
community standards

16. © Copyright Universitätsmedizin Greifswald
BIOINFORMATICS Open Access licences: https://academic.oup.com/journals/pages/open_access/licences; BioModels Licence: https://www.ebi.ac.uk/biomodels/faq#biomodels-licence 16
Reusing other people‘s models
Well-described with metadata & provenance
information; data sources can be linked or
integrated with other data sources.
R1. (Meta)data are richly described with 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
detailed provenance
R1.3. (Meta)data meet domain-relevant
community standards
Example: Models are published with a clear license
information, as are the reference publications

17. © Copyright Universitätsmedizin Greifswald
BiVeS: https://sems.bio.informatik.uni-rostock.de/projects/bives/; Screenshot: FAIRDOMHub: https://fairdomhub.org/models/196 17
Reusing other people‘s models
Well-described with metadata & provenance
information; data sources can be linked or
integrated with other data sources.
R1. (Meta)data are richly described with 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
detailed provenance
R1.3. (Meta)data meet domain-relevant
community standards
Example: Modification of models incl. version
information as provided in FAIRDOMHub.

18. © Copyright Universitätsmedizin Greifswald
Fig.: Curation pipeline for COVID Archives, courtesy Rahuman Sheriff (BioModels); funding: EOSC Fast Track COVID-19; grant no 831644
Example: Making COVID-19 models FAIR

19. © Copyright Universitätsmedizin Greifswald
Fig.: https://healthecco.org/technology/
Example: Making COVID-19 data FAIR
Lea Gütebier
https://healthecco.org/team/

20. © Copyright Universitätsmedizin Greifswald
EU FAIRplus Fellowship Programme: https://fairplus-project.eu/getinvolved/fellowship; SHIP data: https://www2.medizin.uni-greifswald.de/cm/fv/ship/
Picture Gerd Altmann on Pixabay (right) and Jair Lázaro on Unsplash (right)
Example: Making health data FAIR
Esther Thea Inau
0000-0002-8950-2239
Observational
health data

21. © Copyright Universitätsmedizin Greifswald
Photo by Hayley Seibel on Unsplash
“A minimal step towards FAIRness is to provide the data set, as
an entity in its own right, with a PID that is not only intrinsically
persistent, but also persistently linked to the data set (research
object) it identifies. However, without machine-readable
metadata it will still be difficult to find the data, unless one
knows the PID. So a PID is necessary, but not sufficient.”
(https://www.health-ri.nl/fair-principles )
How to: Start
https://combine-org.github.io/events/
Join us at
COMBINE
this year!

22. A little FAIRness is easy to achieve.
Dagmar Waltemath | Department of Medical Informatics
https://twitter.com/waltelab
https://orcid.org/0000-0002-5886-5563