Datasets with bioschemas

Datasets with Bioschemas

Alejandra Gonzalez-Beltran(*), Philippe Rocca-Serra(*), Susanna Sansone(*) and the bioCADDIE Team and Commmunity

(*) Oxford e-Research Centre, University of Oxford

Bioschemas community meeting, Harpenden,Hertfordshire, UK

8th-9th November 2016

The problem

how to describe scientific(*) datasets to enable data discovery

(*) considering in particular

biological and biomedical datasets

Design principles

The model for data description to be designed around the Dataset entity, i.e. a unit of information stored by a data repository:

● Archived experimental datasets which do not change after deposition to the repository; e.g. dbGAP, GEO, ClinicalTrials.org

● Datasets in reference knowledge bases, describing dynamic concepts, such as “genes” whose definition morphs over time; e.g. UniProt

Additionally:

● A dataset and related datasets may available in multiple repositories

● A dataset may be available in multiple forms

Best practices for data on the webhttps://www.w3.org/TR/dwbp

The approach: convergence of metadata elements derived from use cases (top-down) and models/schemas (bottom-up)top-down approach bottom-up approach

(v1.0, v1.1, v2.0, v2.1)

Model serialised as JSON-schemas / JSON-LD with schema.org annotations

Evaluation through Mapping 50+ repositories

Available through DataCite API, mapped to Data citation metadata

Extracting requirements from use cases❖ Selected competency questions

✧ representative set collected from: use cases workshop, white paper, submitted by the community and from NIH and Phil Bourne’s ADDS office

✧ key metadata elements processed: abstracted, color-coded and terms binned binned as Material, Process, Information, Properties; relation identified

Mapping existing metadata schemas/models

❖ schema.org❖ DataCite❖ RIF-CS❖ W3C HCLS dataset descriptions (mapping of many models including DCAT, PROV,

VOID, Dublin Core)❖ Project Open Metadata (used by HealthData.gov is being added in this new

iteration)

❖ ISA❖ BioProject❖ BioSample

❖ MiNIML❖ PRIDE-ml❖ MAGE-tab❖ GA4GH metadata schema❖ SRA xml❖ CDISC SDM / element of BRIDGE model

bottom-up approach

https://biosharing.org/collection/bioCADDIEhttps://github.com/biocaddie/WG3-MetadataSpecifications/

DATS: DAta Tag Suite

Core entities Biomedical extension

Like the JATS (Journal Article Tag Suite) is used by

PubMed to index literature,

a DATS (DatA Tag Suite) is needed for a scalable way to index data sources in the DataMed prototype

https://github.com/biocaddie/WG3-MetadataSpecifications/

Dats - community-driven modelhttps://biocaddie.org/group/working-groups

Dats - community-driven model

DATS core

DATS core and biomedical extension

https://github.com/biocaddie/WG3-MetadataSpecifications/

Mapping dats to schema.org ✧ Missing elements (needed by DATS) submitted to

the tracker; Roughly 80 % of DATS entities and properties can be mapped but alignment is not perfect/less precise), the remaining 20% constitute major gaps

✧

✧ Tracking schema.org and its related Health and Life Science extension evolution (the latter focuses on clinical studies)

https://goo.gl/nNTeW1

https://developers.google.com/search/docs/data-types/datasets

https://developers.google.com/search/docs/data-types/datasets

Adoption and integration

DataMed https://datamed.org/

Dats exported by https://github.com/datacite/spinone/issues/3

● An API endpoint that returns

DataCite metadata in DATS

format is work in progress:

http://api.datacite.org/dats

● DataCite Metadata Schema allows

for a RelatedIdentifier with the HasMetadata relation type this allows linking to the DATS

metadata from a DataCite

metadata record

Martin FennerDataCite

Dats and DCIP supplement mapping

Citation metadata for repositories’ landing page

Mapping between omics-di and DATS

❖ Overlap with DATS

core elements

✧ In red some

DATS extended

elements

Development of a bioschemas specification for datasets

Using the DATS experience for the bioschemas dataset specificationNext step...