Authors: Daniela Cassol (danielac@ucr.edu), Le Zhang (le.zhang001@email.ucr.edu), Thomas Girke (thomas.girke@ucr.edu).
Institution: Institute for Integrative Genome Biology, University of California, Riverside, California, USA.
Overview
Workshop Description
This workshop introduces systemPipe (SP), a generic toolkit for designing and running reproducible data analysis workflows. The environment consists of three major modules implemented as R/Bioconductor packages. systemPipeR (SPR) provides core functionalities for defining workflows, interacting with command-line software, and executing both R and/or command-line software, as well as generating publication-quality analysis reports. systemPipeShiny (SPS) integrates a graphical user interface for managing workflows and visualizing results interactively. systemPipeWorkflow (SPW) offers a collection of pre-configured workflow templates. This hand-on event will include the following topics: (1) brief overview of the design principles and functionalities of the SP toolkit; (2) design and usage of SPR’s command-line interface based on an object-oriented R implementation of CWL; (3) configuration and execution of workflows; (4) construction of custom workflows; (5) configuration and execution of a pre-configured workflow example from start to finish, e.g. RNA-Seq template; (6) parallel execution of workflows on HPC and cloud systems with and without schedulers; (7) generation of technical and scientific analysis reports including visualization; and (8) demonstration of SPS’ core functionalities, the project’s Shiny App.
Pre-requisites
- Basic knowledge of R and usage of Bioconductor packages for NGS analysis
- Basic knowledge of running command-line software
- Basic knowledge of parallelization concepts
Non-essential background reading:
Workshop Participation
Participants will be able to perform all analysis components of this workshop hands-on. Active user participation throughout the event is highly encouraged, including but not limited to lecture material, hands-on sections, and final discussion about package improvements. Participants are encouraged to ask questions at any time during the workshop.
Time outline
1h 45m total
| Activity | Time |
|---|---|
| Overview of systemPipe toolkit | 05m |
| Introduction to SPR’s command-line interface | 15m |
| Showcase RNA-Seq workflow | 20m |
| Configuration and execution of workflows | 20m |
| Generation of technical and scientific analysis reports | 5m |
| Overview of systemPipeShiny core functionalities | 10m |
| systemPipeShiny Showcase | 20m |
| Q&A | 5m |
Workshop goals and objectives
Learning goals
- Recognize the benefits of a generic R-based workflow construction environment that is both scalable and reproducible
- Integration of command-line tools via the CWL community standard
- Rendering of R markdown reports and critical assessment of scientific analysis reports
- Parallelization of big data analysis tasks
Learning objectives
- Identify and practice how to make analysis workflows more robust, reproducible, and portable across heterogeneous computing systems
- Usage of new workflow control class for designing, configuring, and running workflows
- Optimize and debug workflows
- Inspection of technical reports and log files
- Design of new and fully customized workflows
- Practice interactive workflow management and visualization
