GigaSOM.jl - Huge-scale, high-performance flow cytometry clustering in Julia

With GigaSOM.jl, our novel contribution will be allowing the analysis of huge-scale clinical studies, scaling down software limitations. In order to do so, we will implement the parallelization of the FlowSOM algorithm using HPC and increase the maximum number of cells that can be processed simultaneously by the algorithm.

Features

• Analysis and clustering of huge-scale flow cytometry data
• HPC-ready to handle very large datasets
• Load and transform .fcs data files accordingly
• GigaSOM algorithm maps high-dimensional vectors into a lower-dimensional grid
• Automatically determine the required number of cell populations using parallel computing

Check the Background section for some insights on the theory behind our package

See the Functions section for the complete list of documented functions and types.

Contents

• GigaSOM.jl - Huge-scale, high-performance flow cytometry clustering in Julia
• Features
• Contents
• Background
• Introduction
• Multidimensional Data
• Self-organising maps (SOMs)
• Implementation
• How to get started
• Cytometry Data
• Input and output
• Creating a Self Organizing MAP (SOM)
• Functions
• Input
• Process
• Core
• Structs
• Satellites
• Visualisation
• How to contribute to/develop GigaSOM
• Step 1: Retrieve a local version of GigaSOM
• Step 2: Activate GigaSOM