BBMRI-ERIC Biobank Explorer Version 2

Vue CLI 4 application for the biobank explorer; A card detail view on BBMRI-ERIC biobank / collection data

Preparing your MOLGENIS instance

The biobank explorer is built on a specific data model. This is the BBMRI-ERIC model and consists of the following tables:

• eu_bbmri_eric_biobanks
• eu_bbmri_eric_collections
• eu_bbmri_eric_material_types
• eu_bbmri_eric_disease_types
• eu_bbmri_eric_lab_standards

Test model + data can be found in the 'sample-data' folder of this repository. Upload the data via the Advanced data import

Configure the Biobank Explorer

You can configure 2 main settings in the Biobank Explorer.

• Google Analytics
• Negotiator

Enable Google Analytics

We use the Vue Analytics library to connect to Google Analytics. As administrator go to the application settings (/menu/admin/settings/sys_set_app)

Scroll down to Tracking and fill in the fields and add the GA_KEY to:

Google analytics tracking ID

Configuring Negotiator

This version of the biobank explorer is compatible with the Negotiator API in MOLGENIS version 8.3.10.

Disabling filter facets

Facets (except search) can be removed from the list of facets by adding the facet name to an array in the config. This setting can be toggled using the runtimeOptions section in the webpack.prod.conf.js or at runtime using the config entity

runtimeOptions.disabledFilters should contain an array with facet names that should be hidden.

You can find the names here in the filter definitions on github.

"appOptions": {
    ...
    "disabledFilters": ["country"],
    ...
  }

Collection entity (table)

If you want to customize the collection table with additional columns containing an (categorical) mref, you can add these to the view using the following configuration:

"appOptions": {
    ...
    "collectionColumns": [{ "label": "Size:", "column": "order_of_magnitude", "type": "categoricalmref" }],
    ...
  }

Label is the label to show, column is the column name of the collection entity and type is the data type. If you do not specify a type, it will default to a string.

Types

Currently the following types are available:

• mref / categoricalmref (renders as badges)
• string
  • add prefix or suffix to the string like:
    • "prefix": "has"
    • "suffix": "samples"

results in has X samples (where X is the value of the string)

• hyperlink

• range

  • supply in the config:
    • "min": "column_with_min_value" e.g: 'age_low'
    • "max": "column_with_max_value" e.g: 'age_high'
    • "unit": "column_with_unit_value" e.g: 'age_unit'

• object

  • { label: 'Size:', column: 'order_of_magnitude', type: 'object', property: 'size' }

order_of_magnitude query resolves to multiple values as a JSON object, here you can specify which property you want to show.

Optionals

• optionally you can add "display": "badge" to display it as a badge

Default config

You can find the complete default list here or in the runtimeOptions

mind the correctness in the config using " around the key and value

Configuring filter facets for collections

You can remove the standard filters by deleting the filter config. If you deleted them and want them back, check src/config/initialFilterFacets.js

If you want to customize the collection table with additional columns containing an (categorical) mref, you can add these to the view using the following configuration:

"appOptions": {
    ...
    "filterFacets": [
    ...
    {
        "headerClass": facet.headerClass || '',
        "component": facet.component || 'CheckboxFilter',
        "name": "name of the filter as shown in the bookmark, defaults to columnName if ommitted"
        "label": facet.facetTitle || facet.label || facet.columnName,
        "tableName": "table name for the values of the filter",
        "columnName": "column name of the collection table to apply filter to",
        "filterLabelAttribute": "If the source table of the options does not have a label or name, specify column here",
        "initialDisplayItems": "for  multifilters, defaults to 100",
        "maxVisibleOptions": "Maximum amount of options to show per filter before the show more, default 25",
        "showSatisfyAllCheckbox": "If the and / or checkbox is shown ",
        "humanReadableString": "The header for the applied filters sent to the negotiator. EG: Sample(s):",
        "hideFacet": "Should the facet be hidden or not on start.",
    }]
    ...

Contributing

There are 2 ways to test and develop in apps for MOLGENIS.

• locally without MOLGENIS
• locally with MOLGENIS

Test locally without a running MOLGENIS instance

For local testing you can execute the following commands:

# To install the application
yarn install

# To run develop mode
yarn serve

It will render a local version of the core variable catalogue.

Run unit tests

You can run unit tests by executing this command:

# Run once
yarn test:unit

# Run in watch-mode
yarn debug

Run end-to-end tests

You can run end-to-end test locally by running the following command:

yarn test:e2e

Test with a running MOLGENIS instance

For local testing with a running MOLGENIS instance you have to alter the config of the app:

Comment in the following block

config/index.js

module.exports = {
  dev: {

    // Paths
    assetsSubDirectory: 'static',
    assetsPublicPath: '/',
    // Beginning of block
    proxyTable: {
      '/login': {
        target: 'http://localhost:8080'
      },
      '/plugin/directory/export': {
        target: 'http://localhost:8080'
      },
      '/api': {
        target: 'http://localhost:8080'
      }
    },
    // End of block

And comment out this block in the same file.

/**
 * GET and POST interceptors
 * Removes the need for a running backend during development
 */
No mock data available

That is it. Run a molgenis instance on localhost:8080 and start the core variable catalogue with:

yarn dev

Build your MOLGENIS app

See: App developement documentation

You can now create a working application that can be imported in MOLGENIS directly, by executing:

yarn build

You can find the zip-file in the dist/molgenis-app-biobank-explorer_v2.zip.

Customize configuration

See Configuration Reference.

Create a docker image | Molgenis Dev Team

To make a standalone docker image that can be run from the Rancher Cluster perform the following steps:

yarn build:preview

Then build the image with Docker (you have to have Docker running) with the following (tag is required):

docker build -t {tag} .

Then login to the registry

docker login {registry-adress}:{port}

Create a registry tag for the image

docker tag {tag} {registry-adress}:{port}/{imagename}:{optional tag}

Then upload the image to the registry (now you need the tag)

docker push {registry-adress}:{port}/{imagename}:{optional tag}

Now it's available from Rancher.

Go to Rancher, select correct cluster

• workloads > deploy

• Name > name for your workload

• image > {registry-adress}/{imagename}:{optional tag}

• select correct namespace

• Add port > portname: http, publish port: 80, as a: cluster-ip

• env variable: API {molgenis instance you want e.g. molgenis1}

• launch

• workloads > loadbalancing

• add ingress

• website name {description}

• specify hostname: {logicalname}.dev.molgenis.org

• path /, target your created workload, port 80

• save

Done!