Mistic: An open-source multiplexed image t-SNE viewer

Area of Science:

• Computational biology
• Cancer research
• Bioinformatics

Background:

• Tumor microenvironment complexity is crucial for translational research, particularly in immuno-oncology.
• Analyzing multiplexed images from patient samples is vital for understanding tumor ecology.
• Existing tools lack the capability for generating image t-SNEs.

Purpose of the Study:

• To introduce Mistic, an open-source multiplexed image t-SNE viewer.
• To enable simultaneous viewing and analysis of multiple 2D images from patient samples.
• To facilitate exploratory analysis of biomarker expression and cellular phenotypes.

Main Methods:

• Development of Mistic, an open-source software tool.
• Implementation of t-SNE (t-distributed Stochastic Neighbor Embedding) and UMAP (Uniform Manifold Approximation and Projection) for image positioning.
• Provision of multiple layout options for visual preview of multiplexed image datasets.

Main Results:

• Mistic allows simultaneous viewing of multiple 2D images with flexible layout options.
• Image positions can be mapped using t-SNE or UMAP coordinates.
• The tool facilitates identification of images with specific phenotypes and biomarker expression patterns.

Conclusions:

• Mistic provides a novel, freely available solution for generating image t-SNEs.
• The viewer aids in the exploratory analysis of complex multiplexed imaging data.
• This tool supports the selection of relevant images for downstream analysis in cancer research.