GigaSOM.jl: Clustering and visualising massive cytometry datasets

Unleashing the power of GigaSOM.jl to analyse and visualise incredibly large cytometry datasets with remarkable speed

Challenge: Analysing vast cytometry datasets poses a significant computational challenge due to their size and complexity. Existing software packages struggle to handle datasets with millions of cells and dozens of parameters, limiting the ability to extract meaningful insights and explore rare cell populations.

Solution: GigaSOM.jl, a Julia toolkit   developed in a collaboration between the University of Luxembourg and the Luxembourg Institute of Health, revolutionises cytometry data analysis. By leveraging high-performance computing and parallel processing, GigaSOM.jl enables the clustering and visualisation of enormous datasets in minutes. With its ability to load and pre-process Flow Cytometry Standard (FCS) files, distribute data across a network of computation nodes, and generate visualisations, GigaSOM.jl empowers researchers in immunology, cell biology, oncology, and other domains. The toolkit's remarkable speed allows for high clustering resolution, capturing subtle cell distribution patterns and rare populations, ultimately enhancing experimental design and analysis methods in cytometry research. GigaSOM.jl's accessible and flexible implementation paves the way for advancements in big data processing within the field.

Find further information about this example and other similar applications. 

Source:  ULHPC Demonstrators