Skip to content

Latest commit

 

History

History
86 lines (61 loc) · 5.21 KB

README.md

File metadata and controls

86 lines (61 loc) · 5.21 KB
Raw

TraitDB

Build Status DOI

Synopsis

TraitDB is a Ruby on Rails web application for storing and searching trait data. It is in development at NESCent to support working groups.

Installation

TraitDB is a Rails 4 application. It requires ruby and rubygems to run. Other dependencies are specified in the Gemfile. To get up and running with the development environment, you will need Postgres installed. TraitDB can also be configured to work with MySQL if you wish, but Postgres is preferred.

  1. Clone the repository

    git clone [email protected]:NESCent/TraitDB.git

  2. Install dependencies with bundle install

  3. Set your database credentials as environment variables. config/database.yml will read these values out of the environment. If your database server is on a different host, set the host/port as well:

export TRAITDB_PG_DEV_USER="traitdb_dev_user"
export TRAITDB_PG_DEV_PASS="your-password-here"
  1. Run createuser -d -P traitdb_dev_user to create this user in your database.
  2. Run rake db:setup. It will verify rails can connect to the database and run a rake db:migrate to create database tables. If not successful, you can create the databases and users manually, then run rake db:migrate manually.
  3. If you wish to enable Google Sign-in (recommended), you will need to
    • Register an application for Google OAuth 2.0
    • Enable the Google+ API
    • Set the Client ID and Client Secret credentials in your environment:
    export TRAITDB_GOOGLE_APP_ID="your-google-app-id"
export TRAITDB_GOOGLE_APP_SECRET="your-google-app-secret"
  4. Start the server with rails server.
  5. Visit http://localhost:3000 to access the application. You will be shown the about page. If you click Upload, you will be redirected to the sign-in screen. From here, you can sign in with OpenID or a Google Account
  6. Start a delayed_job worker. Delayed job is used to execute dataset imports as a background process. It includes a rake task to start a worker. You can run rake jobs:work in an additional terminal process, or run a worker as a daemon with script/delayed_job start.

Usage

Getting Started - Projects and Users

Data in TraitDB is publicly searchable and organized into projects. Initially there are no projects, and only administrators can create projects. Authentication is handled by OpenID, so in order to get started, you must:

  1. After signing in, there will be an entry in the users table with your email address.
  2. Upgrade this user to an Administrator with the following rake command:
$ rake traitdb:upgrade_admin[[email protected]]
Upgrading [email protected]
  1. Reload your web browser, you will have an Admin menu option.
  2. Click Admin->Projects, and the New Project button.
  3. Fill out the project details and save the new project

Any authenticated user can upload data to any project, but only administrators can create projects and upload Import Configs.

Upload

TraitDB accepts data uploads in CSV format, with a specific focus on data validation and organization. Only logged-in users can upload data. When not logged-in, the upload link will not appear.

In order to upload data into a project, you must write at least one import configuration file in YAML format. This configuration file will contain the project-specific data for your spreadsheets, as well as allowable values and rules for data relationships and which columns to import, ignore, or convert.

For detailed information on writing import configs, see the documentation on the wiki.

Examples for the configuration files are in the lib/traitdb_import directory.

Generally, the CSV files are required to have the following general characteristics

  1. The first row contains column header names The column names include Taxonomic ranks (e.g. Order, Genus, Species), names of traits, and column names for metadata.
  2. Each data row includes trait data and metadata for one Operational Taxonomic Unit (OTU)
  3. Data for a single trait (column) may be either categorical (One or more string tokens separated by a delimeter) or continuous (floating point values)
  4. Source / Reference information for a trait may be in an associated column

As an admin user, you can upload and manage Import Configs for a project. Authenticated users will be able to choose an Import Config when they upload data to the project.

At the upload stage, the user can get information about the Import Config, or download a template CSV file that conforms to it.

Community

For contributing to TraitDB (patches, feature requests, bugs), please see CONTRIBUTING.md. For support requests, please create an issue.

License

TraitDB is open source under the MIT License. See LICENSE.txt for more information.