SPIM workflow manager for HPC.

Area of Science:

• Life Sciences
• Bioinformatics
• Microscopy Image Analysis

Background:

• Life scientists often face challenges with data-intensive processing workflows on High-Performance Computing (HPC) clusters.
• Delegating complex tasks to HPC requires specialized knowledge and can be time-consuming.
• Selective Plane Illumination Microscopy (SPIM) generates large datasets requiring significant computational resources.

Purpose of the Study:

• To introduce a Fiji plugin that leverages the HPC-as-a-Service concept.
• To mitigate the difficulties life scientists encounter when using HPC for data processing.
• To streamline the execution of complex image processing workflows on remote HPC resources.

Main Methods:

• Development of a Fiji plugin integrating HPC-as-a-Service.
• Implementation of data transfer, job management, and results visualization within the Fiji GUI.
• Demonstration using a Selective Plane Illumination Microscopy image processing task.

Main Results:

• The Fiji plugin enables convenient data transfer to remote HPC resources.
• Users can manage pipeline jobs and visualize processed results directly from the Fiji interface.
• Execution on a remote supercomputer improved turnaround time for SPIM image processing, despite data transfer overhead.

Conclusions:

• The developed Fiji plugin significantly simplifies the use of HPC for life scientists.
• This approach enhances the efficiency of data-intensive image processing tasks.
• The tool facilitates faster scientific discovery by reducing computational bottlenecks.