Overview of standards and FAIR, FAIRsharing and the FAIR Cookbook -> Predictive Epigenetics PEP-NET training network, 1 April 2022

1. FAIR services for the life science
Susanna-Assunta Sansone, PhD
Predictive Epigenetics PEP-NET training network, 1 April 2022
Slides: https://www.slideshare.net/SusannaSansone
ORCiD: 0000-0001-5306-5690
Twitter: @SusannaASansone
Professor of Data Readiness
Associate Director, Oxford e-Research Centre
ELIXIR
Interoperability Platform ExCo
elixir-europe.org
Founding
Academic Editor
nature.com/sdata
datareadiness.eng.ox.uk

2. Outline

3. Discoveries are made using shared data and this requires data that are:
• Cited and stored to be discoverable
• Retrievable and structured in standard format(s)
• Richly described to be understandable
Science is about continuity
https://www.forbes.com/sites/gilpress/2016/03/23/data-
preparation-most-time-consuming-least-enjoyable-data-science-
task-survey-says/#276a35e6f637
Data preparation accounts for about 80% of the work of data scientists

4. A set of principles to enhance the
value of all digital resources and
its reuse by humans and machines
FAIR Principles: aspirational guidance

5. DOI: 10.1038/sdata.2016.18
Globally unique and
persistent identifiers
Community defined descriptive
metadata
Community defined
terminologies
Detailed provenance
Terms of access
Terms of
use

6. 80% Metadata
descriptors for the digital objects
20% Identifiers
a sequence of characters that identifies an object

7. • A long-lasting unique reference to a resource resolvable on the internet
• Provides the information required to reliably identify, verify and locate it
• Truly valuable when they are combined and connected
https://orcid.o
rg
Works
People
Organisations
https://ror.org
https://www.doi.org
Funders Projects
https://www.crossref.org https://www.raid.org.au
Persistent identifiers for digital objects
https://blogs.ucl.ac.uk/open-access/2020/07/27/persistent-identifiers-101/

8. http://nometadata.org/logo
Source: https://www.flickr.com/photos/sarahseverson/6245395188
Quote by: Jason Scott, archivist
Metadata - data about the data

9. • Descriptors for a digital object that help to understand what it is, where to find it,
how to access it etc.
• The type of metadata depends also on the digital object
• The depth and breadth of metadata varies according to their purpose
▪ e.g. reproducibility requires richer metadata then citation
Metadata - fundamentals
Illustration by Jørgen Stamp
digitalbevaring.dk CC BY 2.5 Denmark

10. • Domain-level descriptors that are essential for interpretation, verification and
reproducibility of datasets
• The depth and breadth of descriptors vary according to the type of study performed,
generally allowing
▪ experimental components (e.g., design, conditions, parameters),
▪ fundamental biological entities and biomaterial (e.g., samples, genes, cells),
▪ complex concepts (such as bioprocesses, tissues and diseases),
▪ instruments, analytical process and the mathematical models, and
▪ their instantiation in computational simulations (from the molecular level through to whole
populations of individuals)
to be harmonized with respect to structure, format and annotation
Metadata - to describe experiments

11. Metadata make data count!

12. “Most metadata field names and their values are not standardized or controlled”
“Even simple binary or numeric fields are often populated with inadequate
values of different data types”
Standardised description - does it matter? YES

13. 80% Metadata
descriptors for the digital objects
20% Identifiers
a sequence of characters that identifies an object
Interoperability standards
Enable operational processes, such as:
exchange, aggregation, integration, comparison etc

14. Standardization initiated by a
mechanical engineer in 1864...
…...but widely adopted
only a century later!
Interoperability standards – the nuts and bolts

15. Findable
Accessible
Interoperable
Reusable
Providing for a continuum of features,
attributes and behaviours
FAIR Principles: aspirational guidance

16. Outline

17. An intergovernmental organisation that brings
together life science resources
from across Europe, to coordinate them so that
they form a single infrastructure

18. ELIXIR –
a sustainable infrastructure for biological data

19. Specific communities
Human Data, Structural Bioinformatics, Rare
Diseases, Plant Sciences, Microbial Biotechnology ...
FAIR services & resources
Registries, standards, ontologies, identifiers,
data management platforms, stewardship tools,
templates.
Trusted repositories
Deposition databases and portals, scalable
curation, sustainability.
FAIR data techniques
Workflows, reproducible processing, transparent
reporting and provenance, FAIR assessment and
evaluation, FAIRification methods.
FAIR policy and activism
FAIR principles, FAIR leadership & partnering at
the global, European and national level.
FAIR expertise and training
Capability frameworks, skills, data managers
network, training portal.
ELIXIR - in a nutshell

20. Focus on the interoperability platform
Food & Nutrition
+ Toxicology
FAIR services & resources
Registries, standards, ontologies, identifiers,
data management platforms, stewardship tools,
templates.
FAIR data techniques
Workflows, reproducible processing, transparent
reporting and provenance, FAIR assessment and
evaluation, FAIRification methods.

21. FAIR service framework
Ontologies,
formats, reporting
guidelines,
Identifier
Authorities
Metadata
Annotation
& Validation
Citation
Harvesting,
Indexing,
Search
Ontology
Mapping
Identifier
Mapping
Ontology
Lookup
Identifier
resolution
Ontology
Management
Identifier
minting
Standards &
databases
Ontologies Tools Workflows
Identifiers
Registries
Type specific
Mapping &
resolution
Extract-
Transform-Load
APIs
Standards
Services Type specific KBs,
integration &
aggregation
Knowledge

23. An informative and educational resource
FAIRsharing provides curated descriptions and relationship graphs of
standards, databases and policies in all disciplines
COMMUNITY STANDARDS
POLICIES
by funders, journals
and other organizations
DATABASES
including repositories
and knowledgebases
Identifiers
Terminologies Guidelines
Formats

24. Researchers Developers and
curators
Journal
publishers
Societies and
Alliances
Librarians and
Trainers
Funders
Working with and for all stakeholders

25. Guides consumers to discover, select and use these resources with confidence
Helps producers to make their resources more visible, more widely adopted and cited
Total of
over 3595
resources
(March 2022)
repositories
standards
policies
Promoting repositories, standards, policies

26. Search by subject
Powered by our
Subject Ontology of 436 terms
https://fairsharing.org/browse/subject
https://github.com/FAIRsharing/subject-ontology

27. Focus on databases, repositories, knowledgebases
COMMUNITY STANDARDS
POLICIES
by funders, journals
and other organizations
DATABASES
including repositories
and knowledgebases
Identifiers
Terminologies Guidelines
Formats

28. DOI: 10.25504/FAIRsharing.m3jtpg
Example of a
database record

29. License
Maintainer(s)
Standard(s)
Database(s)
Policy(s)
API
Life cycle
status
DOI: 10.25504/FAIRsharing.m3jtpg
At-a-glance-view

30. DOI: 10.25504/FAIRsharing.m3jtpg
Data conditions

31. DOI: 10.25504/FAIRsharing.m3jtpg
Data access points

32. DOI: 10.25504/FAIRsharing.m3jtpg
Related databases,
standards, and policies

33. DOI: 10.25504/FAIRsharing.m3jtpg
Related databases,
standards, and policies

34. Focus on interoperability standards
COMMUNITY STANDARDS
POLICIES
by funders, journals
and other organizations
DATABASES
including repositories
and knowledgebases
Identifiers
Terminologies Guidelines
Formats

35. Identifiers
Terminologies Guidelines
Formats
Conceptual model, conceptual
schema, exchange formats
to represent, contain and
move information
Controlled vocabularies,
thesauri, ontologies
to disambiguate terms and
enable semantic
relationships
Minimum information
reporting requirements,
or checklists
to report the same core,
essential information
Unambiguous, persistent and
context-independent schema
to identify data
and metadata elements
Interoperability standards are the pillars of FAIR
Source:

36. Identifiers
Terminologies Guidelines
Formats
Natural, engineering, humanities & social sciences
816
486
192
19
More than 1500 data and metadata standards
Source:

37. Standard organizations, e.g.: Grass-roots groups, e.g.:
Life and biomedical sciences
Identifiers
Terminologies Guidelines
Formats
486
275
151
8
More than 900 data and metadata standards
Source:

38. Standard organizations, e.g.: Grass-roots groups, e.g.:
• Industry-level standards
• Mostly regulators-driven
• Participation is often regulated
• Standards are sold or licenced
• Formal development process, often
less flexible, could be lengthy
• Charges apply to advanced training or
programmatic access
• Mostly research-level standards
• Open to any interested party
• Volunteering efforts
• Standards are free for use
• Development process varies, more
flexible and adaptable to changes
• Minimal or little funds for carry out the
work, let alone provide training
Understanding their life cycle and landscape
Source:
Identifiers
Terminologies Guidelines
Formats
Formulation
Development
Maintenance

39. DOI: doi.org/10.25504/FAIRsharing.xmmsmr
Example of a
standard record

40. URL: https://fairsharing.org/ISO20691
URL: https://committee.iso.org/standard/68848.html
Relations among
standards

41. URL: https://fairsharing.org/ISO20691

42. Tracking the evolution of standards

43. Translational Medicine
Clinical Developments
URL: beta.fairsharing.org/PistoiaAllianceFIPs
(work in progress!)
A collaboration with their
FAIR Implementation WG
Disclaimer: These profiles speak for a limited community and do not represent any company standards
Building and comparing
“FAIR profiles”

44. Clinical Developments
Disclaimer: These profiles speak for a limited community and do not represent any company standards
Snapshot of the
semantic and syntactic
standards used
A collaboration with their
FAIR Implementation WG

45. URL: fairsharing.org/graph/3513
These are profiles:
Of the organizations and their RIs, with
their data resources and standards
EOSC-Life “FAIR profiles”

46. Connected to the data resources and
standards they are associated with
Organizations and users

47. Adopters and collaborators include:
An endorsed output of the
FAIRsharing WG
(since 2015):
A WG (since 2015) in:
Researchers in academia,
industry and government
Developers & curators of
resources and tools
Research data facilitators,
librarians, trainers
Society, unions
and community alliances
Journal publishers and
organisations with data
policies
Funders and data
policy makers
A recommended resource in EOSC reports
Used by all stakeholder groups
https://fairsharing.org/communities
A de facto element of the EOSC ecosystem

48. Stakeholder Advisors
● Amye Kenall, VP of Publishing and Product, Research Square
● Adam Leary, Oxford University Press
● Catriona MacCallum, Hindawi
● Dagmar Meyer, European Research Council, Executive Agency
● Dominic Fripp, JISC, UK
● Emma Ganley, Protocols.io
● Geraldine Clement-Stoneham, Medical Research Council
● Helena Cousijn, DataCite
● Iain Hrynaszkiewicz, PLoS
● Imma Subirats, FAO of the United Nations
● Kiera McNiece, Cambridge University Press
● Luiz Olavo Bonino, GO-FAIR
● Marina Soares E Silva and Sarah Callaghan, Elsevier
● Michael Ball, Biotechnology and Biological Sciences Research Council
● Mike Huerta, NIH National Library of Medicine
● Molly Cranston and Guillaume Wright, F1000Research
● Nick Everitt and Matthew Cannon, Taylor and Francis
● Scott Edmunds, GigaScience, Oxford University Press
● Simon Hodson, CODATA
● Theo Bloom, British Medical Journal
● Thomas Lemberger, EMBO Press
● Wei-Mun Chan, eLife
● Sowmya Swaminathan, Springer Nature
Current operational Team
● Allyson Lister, Content and Community Lead
● Milo Thurston, Technical Lead
● Ramon Granell, Data Enrichment & Quality Manager
● Delphine Dauga, Data Curator Manager
● Hiring in progress, Web Developer
● Dominique Batista, Research Software Engineer
● Philippe Rocca-Serra, Co-Founder
● Susanna-Assunta Sansone, PI and Founder
● and many collaborators and contributors!
Executive Advisors
● Varsha Khodiyar, HDRUK
● David Carr, Independent expert
● Chris Graf, Springer Nature
● Marta Teperek, Data Stewardship Coordinator, TUDelft
● Robert Hanisch, Director, NIST Office of Data & Informatics
● Peter McQuilton, FAIRsharing Founding Member, GSK

49. Outline

50. Beyond the hype
Large body of generic FAIR
guidance
Motivations
Non-specific guidance for
the life sciences
Ambitions
Target specific situations to deliver a guide with
applied examples
Join academia and industry forces to make the
case for FAIR data management
Build capacity for high quality data
management in the private and public sectors

51. 51
Different contexts mandate different standardization strategies
Molecular data
Clinical (observation based)
data
Clinical trial (event based) data
FAIRification paths: one size does not fit all

52. Molecular data
Selecting a ‘standard stacks’ for the FAIRification
Terminologies
Guidelines
Formats

53. FAIR Cookbook:
turning knowledge into recipes
What is it?
An online, ‘live’ resource
for the life sciences
A collection of recipes
that cover the operation
steps of FAIR data
management
Who is it for?
Who developed it?
Researchers and data
managers professionals
in the life sciences, from
academia and industry
Including ELIXIR
members
faircookbook.elixir-europe.org

54. FAIR Cookbook: learning objectives
Learn how to improve the FAIRness with exemplar datasets
Understand the levels and indicators of FAIRness
Discover open source technologies, tools and services
Find out the required skills
Acknowledge the challenges
faircookbook.elixir-europe.org

55. Recipes that cover all aspects of FAIRness
Over 6o recipes (March
2022) released and
many more in progress!
Covering technical
processes, with
examples in the life
sciences, including
omics, pre-clinical and
clinical areas
But not limited to it!

56. Anatomy of a recipe

57. https://doi.org/10.1038/s41597-019-0286-0
Applied examples: omics data matrices

58. Citability and credit to the authors

59. We are working to tag the recipes with a
‘dataset maturity model’
It show the level of FAIRness you can reach by
applying a specific recipe to improve a dataset
https://fairplus.github.io/Data-Maturity
Maturity level and indicators of FAIRness

60. The capability maturity model - the ontology example
Which capabilities are needed to
improve the semantic
understanding of my data?
The optimum level of FAIRness
is a trade-off between desired
data reuse level and cost to
achieve that level!
No use of
ontologies
Use of internal
ontologies
Use of
community
ontologies
+ Ontology service to
manage several
ontologies, mapping,
versioning etc.
+ Term suggestion,
automatic annotation,
terms conflict
resolution etc.

61. No use of
ontologies
Use of internal
ontologies
Use of
community
ontologies
+ Ontology service to
manage several
ontologies, mapping,
versioning etc.
+ Term suggestion,
automatic annotation,
terms conflict
resolution etc.
The capability maturity model - the ontology example
A dedicated (set of)
recipe will help to move
from Repeatable to
Defined level

62. Almost 100 life sciences professionals, researchers and data managers
FARIplus
partners
Industry
+
Academia
Creators and contributors
ELIXIR
Nodes
represented

64. Watch the webinar for more information,
and watch out more recipes!
elixir-europe.org/events/fairplus-
webinar-discovering-fair-cookbook
faircookbook.elixir-europe.org
fairplus-cookbook@elixir-europe.org

65. FAIRification is a team sport,
it takes a village!
It is not more optional,
but it is work in progress……