SushiRunner

A Dash interface to submit Sushi applications via B-Fabric using SushiFabric and WorkflowManager.

Report Bug · Request Feature

Note: This project is built using the bfabric-web-apps library and was forked from the bfabric-web-app-template.

---

About

SushiRunner is a web-based interface for submitting Sushi applications using B-Fabric datasets as input. It bridges the B-Fabric LIMS system with the SushiFabric command-line interface and the WorkflowManager execution engine.

This app allows users to:

• Load dataset metadata from B-Fabric
• Configure Sushi application parameters through a Dash-based UI
• Generate the required dataset.tsv and parameters.tsv files
• Submit jobs to the Sushi infrastructure without needing to manually write scripts

This app is intended as a proof-of-concept, showcasing generalized support for Sushi-based execution workflows directly from the B-Fabric environment.

---

Key Features

• Dash UI for selecting and configuring Sushi apps
• Generic + app-specific parameter forms (RAM, cores, options, etc.)
• TSV generation for Sushi input
• Seamless B-Fabric authentication and dataset loading
• Job submission via sushi_fabric (no Redis dependency)
• Optional project charging and reporting integration

---

Workflow Overview

1. User launches app from a B-Fabric dataset page
2. Dataset metadata is automatically loaded
3. User configures the job parameters
4. dataset.tsv and parameters.tsv are generated
5. sushi_fabric is invoked with the appropriate class and args
6. WorkflowManager handles execution, monitoring, and completion emails
7. Workunits and logs are updated in B-Fabric

---

Example Applications

Each Sushi app is defined as a standalone Dash module. For example:

• FastqcApp supports:
  • Generic inputs: RAM, cores, partition, etc.
  • App-specific toggles: paired, showNativeReports, etc.
  • Table-based dataset preview
  • Dynamic job submission via run_main_job

Other supported applications follow a similar structure and can be added modularly.

---

Useful Resources

• Sushi Documentation
• Sushi App Example - FastQC
• SushiFabric
• WorkflowManager
• EZRun R Package
• FGCZ Sushi Job Monitor

---

Built With

• Dash
• Flask
• Plotly
• bfabric-web-apps
• SushiFabric
• WorkflowManager

---

Quickstart

1. Clone the Repository

git clone https://github.com/GWCustom/SushiRunner.git
cd SushiRunner

2. Create and Activate a Virtual Environment

Using virtualenv (Linux/Mac):

python3 -m venv venv
source venv/bin/activate

Using virtualenv (Windows):

python -m venv venv
venv\Scripts\activate

Or using conda:

conda create -n sushi-runner pip
conda activate sushi-runner

3. Install Dependencies

pip install -r requirements.txt

4. Set Up .bfabricpy.yml

Create this file in your home directory (~/.bfabricpy.yml):

GENERAL:
  default_config: PRODUCTION

PRODUCTION:
  login: your_username
  password: your_password
  base_url: https://your-bfabric-api-endpoint

5. Run the App

python3 index.py

Visit http://localhost:8050 in your browser.

---

License

Distributed under the MIT License. See LICENSE for details.

---

Contact

GWC GmbH – GitHub
Griffin White – LinkedIn
Marc Zuber – LinkedIn