This tutorial will introduce you to Bioschemas, and will present how to include schema.org markup to make your resource(s) more discoverable on the Web. At the end of the session, attendees will be able to 1) Understand what is Bioschemas and how to use it 2) Have examples of deployments using Bioschemas

1. Improving discoverability for Life
Sciences resources
Alasdair J.G. Gray
Bioschemas Leadership Team Chair
Heriot-Watt University/Elixir-UK
Bioschemas
ELIXIR All Hands Tutorial
Lisbon, Portugal – 19 June 2019

2. Google Search
2http://bioschemas.org

3. Google Search
3http://bioschemas.org

4. Google Dataset Search (Sept 2018)
4
https://toolbox.google.com/datasetsearch
http://bioschemas.org
https://www.blog.google/products/search/making-it-easier-discover-datasets/

5. Picture: Carole Goble, Turing Lecture 2018
Schema.org: Semantic Markup for the Web

6. Structured data → descriptors
● Types
(614)
What we can
say about
those things
● Properties
(905)
What we are
talking about

7. Bioschemas
• Community initiative built on top of
schema.org
• Aim
• Improve data discoverability and
interoperability in Life Sciences
• Approach
• Add Life Science types to schema.org
• Provide usage guidelines and examples
• 6 Minimal properties
• Link to domain ontologies
• Support software
Profile over schema.org
Layer of constraints + documentation +
extensions Specification
Data model
Minimum information
Controlled vocabularies
Cardinality
Documentation
Examples
New (properties | types)

8. Findable Accessible Interoperable Reusable
★Globally unique
identifiers
★Community
defined enriched
metadata
★Indexable by
search engines
★JSON-LD/RDFa
★Link to
controlled
vocabularies
★Links to other
resources
★ License
★ Provenance
★Retrievable
★HTTP

9. Schema.org for Datasets
Schema definition:
●Dataset: A body of structured
information describing some
topic(s) of interest
http://schema.org/Dataset
●91 properties including:
○name
○description
○isFamilyFriendly
9

10. Google Dataset Profile
• 2 required properties
• Used for Google Dataset Search
• 10 recommended properties
• Link to DataCatalog
• Link to DataDownload
Other profiles: Events, Jobs,
...
https://developers.google.com/search/docs/data-types/dataset
Google Dataset Profile

11. Compliant with Google
Dataset Profile
• 5 minimal properties
• 8 recommended properties
• Link to DataCatalog
• Link to DataDownload
http://bioschemas.org/specifications/Dataset/
Bioschemas Dataset Profile

12. Extending Schema.org for the Life Sciences
7 release candidates
Submission in progress!

13. More types in development

14. 14
Profile Version Group Live Deploys Status notes
DataCatalog 0.2 (Jun 2019) Data Repos 20 0.2 fixes minor issues
Dataset 0.3 (Jun 2019) Datasets 23 0.3 fixes minor issues
Event 0.1 (July 2018) Events 7 Used by TeSS: undergoing revision due to addition of CourseInstance
Sample 0.2 (Nov 2018) Samples 1
Taxon 0.3 (Nov 2018) Biodiversity 0
Tool 0.1 (Mar 2018) Tools 5 0.3-DRAFT based on bio.tools profile, needs review
TrainingMaterial 0.2 (July 2018) Training 0 Used by TeSS: 0.5-DRAFT incorporating changes from Course
Current Bioschemas Profiles

15. Draft Bioschemas Profiles
15
● Beacon: 0.2-DRAFT 2018-04-23
● BioSample: 0.1-DRAFT
● ChemicalSubstance: 0.2-DRAFT 2019-06-11
● Course: 0.6-DRAFT 2019-06-06
● CourseInstance: 0.6-DRAFT 2019-06-06
● DNA: 0.1-DRAFT 2018-11-13
● DataRecord: 0.2-DRAFT 2019-06-14
● Gene: 0.5-DRAFT 2019-06-14
● Journal: 0.1-DRAFT 2019-02-08
● LabProtocol: 0.3-DRAFT 2019-06-14
● MolecularEntity: 0.2-DRAFT 2019-11-15
● Organization: 0.1-DRAFT 2018-03-13
● Person: 0.1-DRAFT 2018-03-14
● Phenotype: 0.1-DRAFT 2018-11-15
● Protein: 0.8-DRAFT 2019-05-08
● ProteinAnnotation: 0.4-DRAFT 2018-02-25
● ProteinStructure: 0.5-DRAFT 2018-08-15
● PublicationIssue: 0.1-DRAFT 2019-02-08
● PublicationVolume: 0.1-DRAFT 2019-02-08
● ScholarlyArticle: 0.1-DRAFT 2019-02-08
● SemanticAnnotation: 0.1-DRAFT 2019-02-08
● Standard: 0.1-DRAFT 2018-01-01
● Study: 0.1-DRAFT 2018-11-15
● Tool: 0.3-DRAFT 2018-11-21
● TrainingMaterial: 0.6-DRAFT 2019-06-06
● Workflow: 0.1-DRAFT 2019-02-08

17. Mapping ProfileUse cases
Mockup
Adoption
Testing Application
Profile Creation Process

18. Bioschemas Software
29 November 2018 http://bioschemas.org 19
Bioschemas Generator
● Supports all profiles
○ Current and draft
● Validates input
● Form generated from
YAML description
● Examples extracted from
profile

19. Exploiting Bioschemas Markup

20. TeSS: Specialised Search
http://bioschemas.org
• contact
• description
• endDate
• eventType
• hostInstitution
• location
• name
• startDate
• …
Bioschemas Event:
29 November 2018 21

21. http://bioschemas.org
• description
• keywords
• name
• provider
• url
Bioschema DataCatalog:
• alternateName
• citation
• dateCreated
• licence
• …
Automated Data Curation

22. Data Exchange: Without an API
MarRef → BioSamples
https://github.com/EBIBioSamples/bioschemas_marref_demo/blob/master/Summary.md

23. BKG Explorer
Built over Bioschemas markup crawled from 30 live deployments
20,000 pages

24. Bioschemas
What?
• Exploiting schema.org to make Life Sciences
resources more discoverable
• Search engines will index and understand
markup
How?
• Extending schema.org vocabulary for life
sciences
• 7 release candidate types
• Provide guidelines on how to markup
resources

25. 200+
People
7
Tutorial
s
(2018)
17Type
s
6Publications
(2018)30Profiles
62Sites
11M+
Pages
Bioschemas Community
http://bioschemas.org/liveDeploys
http://bioschemas.org/
liveDeploys

26. http://bioschemas.org
Acknowledgements http://bioschemas.org/people

27. http://bioschemas.org/ @bioschemas https://github.com/bioschemas/
Join Bioschemas: http://bioschemas.org/howtojoin/

28. Creating and Deploying
Bioschemas Markup
Material from: Justin Clark-Casey
License: Attribution 4.0 International (CC BY 4.0)
Kenneth McLeod

29. Creating Bioschemas markup
● Markup is in a format called JSON-LD
● Embedded directly into webpages
● Let’s look at an example of the DataCatalog schema as used by Bioschemas
○ This comes from schema.org but Bioschemas adds
■ Mandatory/recommended/optional properties
■ Cardinality constraints

31. Markup can be placed in either the
head or the body.

32. Let’s look at this in Google’s Structued
Data Testing Tool

35. @context is overwritten by Google
Technically any prefixes can be defined here, e.g.,
"@context":["https://schema.org", {"OBI":"http://purl.obolibrary.org/obo/OBI_" ...}],
"@type":["Sample","OBI:0000747"] …
BUT, Google will overwrite this with the basic "@context": "http://schema.org"

36. @id - gives a node a URL
Without @id there are auto-generated URLs for nodes, e.g.,
<script type="application/ld+json">{
"@context" : "https://schema.org",
"@type" : "DataCatalog", ...
becomes:
_:genid2d4335ed7c72694275bea5b6a86ad9f82b2db0
<http://www.w3.org/1999/02/22-rdf-syntax-ns#type>
<https://schema.org/DataCatalog> .
Bad for Linked Data as no one can reference this.

37. @id - gives a node a URL
With an@id you choose the URL for nodes, e.g.,
<script type="application/ld+json">{
"@context" : "https://schema.org",
"@type" : "DataCatalog",
"@id" : "https://www.ebi.ac.uk/biosamples" …
becomes:
<https://www.ebi.ac.uk/biosamples> <http://www.w3.org/1999/02/22-rdf-syntax-
ns#type> <https://schema.org/DataCatalog> .

38. Warning! Don’t use the same @id for everything
DataCatalog & Dataset defined separately, but combined into a single entity:

39. GSDTT
common errors
If you don’t meet
Google’s
desired property
specification for
a given type you see
errors like:
If Bioschemas spec says this is OK, you can
ignore error (FYI it is a real error)
Not min properties in
Bioschemas; do what you
want
This error is caused by the
incorrect target type of location.
Description is min property
for Bioschemas (ie
mandatory)

40. Bioschema’s Types not yet accepted by Schema:
Ignore these

41. Markup Generator

42. Example:
https://bio.tools/blast

43. https://blast.ncbi.nlm.nih.gov/Blast.cgi

44. https://bioschemas.org/devSpecs/Tool/

45. Evolving Best Practices
● At the moment we largely create markup by hand with validation through
Google’s testing tool
○ More validators and tools on the way, see bioschemas.org/tools
● Make pages with markup reachable from your sitemap.xml
○ This will make it easier for some applications to find it.
● Avoid adding Bioschemas markup to the page dynamically (e.g. through
Javascript)
○ Applications trying to find your data may not have the resources to render pages.
● Specify an @id
● Evolving guidance at
https://github.com/BioSchemas/specifications/wiki/Technical

46. Questions?
● bioschemas.org
● bioschemas.org/groups/Technical
● https://bioschemas.org/software/
● Google Structured Data Testing Tool
● kcm1@hw.ac.uk