Captured Metadata¶

This section provides an overview on what metadata elements are captured with the GHGA Metadata Schema.

A breakdown of each metadata element described in the different entities will provide more insight on what elements are required for the functionality of GHGA, mandatory properties and recommended or optional information that can be provided by the data submitters.

Research Metadata¶

The Research Metadata focuses on the reusability and FAIRness of the data. The Research Metadata encapsulates a set of classes that shape the model in accordance to the performed experiment ensuring reproducibility and scalability. The classes include Individual, Biospecimen/Sample, Experiment, Experiment Method, Analysis and Analysis Method.

Individual¶

The Individual entity collects non-personal metadata about the Individual from which the samples were collected. The content of the Individual entity is crucial to identify cohorts of interest and gives valuable insight on the target group of an experiment.

Individual metadata properties¶

In order to describe an Individual, data submitters are required to provide information about sex and are recommended to provide information about phenotypic features and diagnoses. Additional information about the geographical region and ancestries can also be provided.

The Individual entity is linked to the Sample entity.

Individual Supporting File¶

If the data submitters have additional information to provide about the Individual, then they can make use of the Individual Supporting File class.

Individual Supporting File metadata properties¶

The Individual Supporting File requires the following information: file format, file name, dataset associated with this file as well as if the file is included as part of the submission.

Sample¶

The Sample describes the extracted sample and tissue for an experimental process. Furthermore, the Sample is defined as a limited quantity of something to be used for testing, analysis, inspection, investigation, demonstration, or trial use. A sample is prepared from a biospecimen (isolate or tissue). The Sample entity includes slots describing the Biospecimen from which the sample was prepared, which are highlighted with a "biospecimen_" -prefix. The Biospecimen is defined as any natural material taken from a biological entity for testing, diagnostics, treatment or research purposes.

The Sample is linked to the Individual entity describing the collection of biological material. Further, it is linked to the Experiment.

Sample metadata properties¶

The Sample entity requires data submitters to provide the name, description, whether the sample is case or control, as well as the alias to the Individual from which the Sample was derived. In addition to the mandatory information, the data submitter is recommended to provide information about the storage, whether the sample can be classified as diseased or healthy, and biological replicate information. Other optional fields include the type of the sample and an EGA accession ID.

To further describe the Biospecimen from which the sample was prepared, data submitters are required to provide information about the age of the individual at the time of sampling and the biospecimen tissue. The recommended properties include a name for the biospecimen, its type, the vital status of the individual at the time of sampling, and the biospecimen isolation and storage.

Experiment¶

An Experiment is an investigation that consists of a coordinated set of actions and observations designed to generate data with the goal of verifying, falsifying, or establishing the validity of a hypothesis. The Experiment entity describes the experimental setup or a protocol used for performing an omics experiment.

The Experiment entity is linked to the Sample which implies one or more samples associated with this Experiment and the Experiment Method which implies one or more experiment methods associated with this Experiment. The outcome of an Experiment results in Research Data File.

Experiment metadata properties¶

The Experiment entity requires the data submitters to provide information about experiment method, title, description and sample information.

Experiment Method¶

The Experiment Method entity captures information about the protocol that was followed to perform the omics experiment.

Experiment Method metadata properties¶

The data submitter is required to provide information about the name, description, library type, selection methods, preparation, the instrument model and sequencing layout to fulfil the Experiment Method pertaining to an omics experiment. Furthermore, it is recommended to provide information about the library preparation kit name and manufacturer, the primer, end bias, flow cell ID and type, rna seq strandedness, sample barcode read and sequencing center. The optional properties include target region, target coverage and EGA accession ID.

Experiment Method Supporting File¶

Any additional information pertaining to the Experiment Method can be provided as Experiment Method Supporting File.

Experiment Method Supporting File metadata properties¶

The submitter is required to provide information about the file format, the Experiment Method this file supports, the file name, the Dataset associated with this file and whether this file is included as part of the submission.

Research Data File¶

The Research Data File entity is generated as a result of a direct measurement or sequencing of a sample. They are the raw files that were generated while carrying out an Experiment. This entity is linked to the Experiment that produced this file as well as the Dataset that is associated with this file. The GHGA metadata model emphasizes on the submission of a Research Data File alone or accompanied by Process Data Files for enabling reconstruction of the analytical process representing a submission.

Research Data File metadata properties¶

The Research Data File entity requires a data submitter to provide information about the file format, technical replicate, the Experiment that generated this Research Data File, the file name, the Dataset and whether the file is included in the submission. It is also recommended to provide information about the sequencing lane. The EGA accession ID is an optional property.

Analysis¶

The Analysis entity permutates a Research Data File into a Process Data File through a computational workflow.

The Analysis entity is linked to Experiment via Research Data File and also linked to Analysis Method.

Analysis metadata properties¶

The data submitter is required to provide information about the Analysis title, the Analysis Method that is associated with this Analysis and the Research Data Files that this Analysis used as input to generate Process Data Files.

Analysis Method¶

The Analysis Method entity describes the computational workflow that created Process Data Files in more detail.

Analysis Method metadata properties¶

For this submission, the data submitter is required to provide information about the Analysis Method name, a description, type, the workflow name, workflow repository and workflow DOI. Furthermore, it is also recommended to provide information about the workflow version, workflow tasks, parameters and software versions.

Analysis Method Supporting File¶

For any additional information pertaining to the Analysis Method entity, the data submitter can use the Analysis Method Supporting File.

Analysis Method Supporting File properties¶

For supporting information about the Analysis Method, the data submitter is required to provide information about the file format, the Analysis Method this file supports, the file name, and the Dataset associated with this file.

Process Data File¶

The Process Data File entity is a generated outcome of an analysis, alignment or processing of a Research Data File. The Process Data File entity is linked to the Dataset that is associated with the file and the Analysis that produced this Process Data File.

Process Data File metadata properties¶

The Process Data File entity requires the data submitter to provide information about the file format, the Analysis from which this file originated, the file name, and the Dataset. Additionally, it is also possible to submit an optional EGA accession ID.

The submission process allows submission of processed data file alone but this is discouraged as it requires creation of empty Experiment class.

Administrative Metadata¶

The Administrative Metadata focuses on managing the resources such as creation or acquisition of the data, rights management, and disposition.

Dataset¶

GHGA presents its content to potential data requesters and submitters with the Dataset entity, which focuses on sharing functionality by describing the contents at a high level. Each Dataset is linked to a Data Access Policy, which builds the legal basis for the sharing of data. One Dataset has links to Experiment via Research Data File, and to Analysis via Process Data File. The Dataset is also indirectly linked to Analysis Method, Experiment Method and the Individual via Analysis Method Supporing File, Experiment Method Supporing File and Individual Supporting File, respectively. The Dataset entity is also linked to Study.

Dataset metadata properties¶

The Dataset entity is aimed at capturing relevant information about a dataset itself. The mandatory information required for Dataset entity includes title, description, type, the Data Access Policy that regulates this Dataset and the Study. Optionally, the data submitter could also provide information about an EGA accession.

Data Access Policy and Committee¶

Depositing data at GHGA requires a data submitter to provide a Data Access Committee (DAC) and Data Access Policy (DAP). This ensures controlled access to their deposited data and a clear guideline for data requesters to access the data. This includes a defined contact person and a consent-based legal basis for getting access to a dataset. The DAP entity is linked to Dataset and DAC.

DAP and DAC metadata properties¶

A DAP is directly linked to the DAC and Dataset entity, thus providing the condition under which the data deposited at GHGA can be re-used by a data requester. The submitter must provide an alias, name, description and either the policy text for the DAP or the URL where the DAP is stored, data use permission and modifier using the Data Use Ontology, and the DAC that governs this DAP.

A DAC is linked directly to DAP and requires information about the DAC email and institute and optionally about an EGA accession.

Study¶

All data deposited at GHGA is subject to a specific study, under which relevant data has been aggregated. A Study is an experimental investigation of a particular phenomenon and involves a detailed examination and analysis of a subject to learn more about the phenomenon being studied. A detailed description of a study can guide data requesters to identify the most relevant datasets for their own research. The Study entity is linked to Dataset and optionally to Publication.

Study metadata properties¶

In order to describe a Study, data submitters are required to provide information about the study affiliation(s), the title, a description and the study type. Optionally, they can provide an EGA accession ID.

Publication¶

The Publication entity refers to a journal article, book, or conference paper that presents original research or scholarly findings in a specific field of study, contributing to the collective knowledge and understanding of that discipline.

The Publication entity is linked to Study.

Publication metadata properties¶

The data submitter is required to provide information about the Study associated with this publication. Optionally, the data submitter can also provide information about the publication title, abstract, author, year, journal and DOI.

Submission Spreadsheet¶

The Submission Spreadsheet for GHGA Archive captures the above-mentioned metadata in an ordered fashion. The submission process is tailored to replicate a “front-to-end” workflow:

Which individuals are subject to investigation?
Which biological material was collected and how?
Which experiment was performed and how? Which files were generated as result?
Which analysis was performed and how? Which output files were generated from which input files?
How can this data be shared in meaningful datasets? Which conditions apply and who governs them?

The submitter can also make use of the supporting files to enrich their data with additional information, such as experimental protocols for the Experiment class or workflow parameter files for the Analysis class. The supporting files can contain structured metadata, such as Phenopackets or PED files for the Individual class. Since supplementary files are encrypted and inaccessible without DAC approval, this allows submissions of clinical or personal metadata that can only be accessed by requesters after signing a Data Transfer Agreement. Hence, the metadata model can only indicate the presence of supplementary files for classes and signify that a submission contains additional information but not process their content. This results in a submission process, which will appear much more natural to submitters, while maintaining a high degree of flexibility to adapt to different omics types.