Skip to content

Latest commit

 

History

History
143 lines (88 loc) · 8.43 KB

documents.rst

File metadata and controls

143 lines (88 loc) · 8.43 KB

Documents

Mica handles several type of documents, specific to the epidemiological studies domain: network, study, datasets etc. These document types have their own internal structure (to allow relationships between them and to ensure basic search), but can also be extended with custom fields. The default set of fields is the one promoted by Maelstrom Research. This default description model should fit with your needs in most of the cases.

All the documents follow the :doc:`publication-flow` except the :ref:`data-access-requests` (which is a form privately exchanged between a researcher and the study/consortium).

Types

Network

A network is a group of epidemiological studies that has specific research interests. It is described using the following fields: name, aims, investigators, contact information and participating studies. It can also be related to other networks.

Individual Study

An individual study is defined as any epidemiological study (e.g. cohort, case control, cross sectional, etc.) conducted to better understand the distribution and determinants of health and disease. It is described using the following fields: name, objectives, investigators, contact information, design, data collection timeline, target number and characteristics of participants, and related scientific publications and documents. A study can include one or more populations described below.

Population

A population is a set of individuals sharing the same selection criteria for enrollment in a study. It is described using the following fields: name, sources of recruitment, participant characteristics, and number of participants. A population is linked to one or more data collection events according to the number of follow-ups.

Data Collection Event

A data collection event is a collection of information on one or more population(s) over a specific period of time (e.g. baseline, follow-up 1, follow-up 2). It is described using the following fields: name, start and end date, and data sources (e.g. questionnaires, physical measures, biosample measures, etc). A data collection event may be associated to one or more populations and it can include one or more datasets.

Harmonization Initiative

A harmonization initiative is defined as a research project harmonizing data across individual studies to answer specific reseach questions. It is described using the following fields: acronym, contact information, objectives, start year, end year and related documents. A harmonization initiative can include one or more harmonization protocols (dataschema).

Collected Dataset

A collected (study-specific) dataset holds metadata about the variables collected within a data collection event. The metadata is described using a standardized format of data dictionary which provides information on collected variables’ definitions and characteristics (e.g. type, unit, categories, and area of information covered). It can be associated to a study by specifying a data collection event.

Collected Variable

A collected variable is a variable that was collected, measured, or constructed within a study protocol. It is described using the following fields: name, label, description, type, unit, categories, and area of information covered. If the collected dataset includes data, summary statistics of the collected variable can be published on the web portal (e.g. means, minimum, maximum, counts and percentages). Each collected variable is part of one and only one study-specific dataset.

Harmonization Protocol

A harmonization protocol holds metadata about core variables constructed from multiple collected datasets. The metadata is described using a standardized format of data dictionary which provides information on harmonized variables’ definitions and characteristics (e.g. type, unit, categories, and area of information covered): this represent the data schema of the harmonization protocol. It can be optionally associated to the harmonized data.

Data Schema Variable

A data schema variable is the harmonization protocol reference variable. Each harmonized variable will implement a corresponding data schema variable.

Harmonized Variable

A harmonized variable is a core variable (common format) generated by multiple individual studies. It is described using the following fields: name, label, description, type, unit, categories, and area of information covered. If the harmonization protocol includes data, summary statistics of the harmonized variable (e.g. means, minimum, maximum, counts and percentages) can be published on the web portal. Each harmonized variable is part of one and only one harmonization protocol.

Research Project

A research project reports information about the work that was conducted thanks to the network/study data: research objectives and results, contact information, status timeline. It could be somehow related to a data access request but not necessarily.

Data Access Request

A data access request is an application form that researchers submit in order to gain access to the network/study data. There is a predefined workflow from submission to approval described in more detail in :doc:`web-user-guide/data-access-request`.

Data Access Amendment

A data access amendment is an application form that researchers submit to request changes to a pre-approved data access request. The data access amendment workflow is identical to that of data access requests.

Search

Mica search engine allows to look into the domain while applying criteria on each type of document. The result of this combined query can be of any type. For example:

  • search for variables about alcohol, associated to studies having collected biosamples, and being part of a network
  • search all studies having collected biosamples and having variables about alcohol, and being part of a network
  • ...

Associations

The following diagram describes the various documents that can be published in the Mica web portal. Each of them can be edited individually in the Mica Web Application administration interface (except variables, defined in the Opal servers).

images/mica-associations.png

Annotations

Individual studies can be annotated in two ways:

  • Manually annotated by a taxonomy vocabulary.
  • Inferred annotations from their associated variables.

Manual annotations by vocabulary are primarily for studies that do not have any associated variables but need to be classified by a taxonomy vocabulary.

Permissions

Three types of permissions can be granted to a user. Each permission is defined by a user role each of which applies different level of restrictions on a document. The table below lists each role and corresponding restrictions:

Role Description
Reader Read-only access to the document in draft mode with its revisions and its associated files.
Editor Edit access to the document in draft mode with its revisions and its associated files. Publication or permanent deletion are not permitted.
Reviewer Full access to the document, including its publication, permanent deletion and permissions.

Revision History

The revision history of a document is the succession of states after each edition (state refers to the content of the document, not its status). This history of changes allows to:

  • view changes,
  • reinstate a revision,
  • identify which state is published.

Comments

To enhance the collaboration between Mica users, each member can add a comment on any Mica domain document as well as data access requests documents. Mica can be configured to send email notifications when a comment is added or updated.

Cart

The cart groups a set of variables based on one or more search criteria. User can edit the cart by adding or removing variables, download its content (variables) as a CSV file and use it as a search criterion. Variables of a cart can be attached to a Data Access Request.

The cart can also contain studies and networks. Similarly the content of the cart can be downloaded as CSV files (for internal content analysis). Documents of the cart also be compared between each other.