SUFI

SUFI performs spectral unmixing of multispectral fluorescence images to extract fluorophore-specific signals for quantitative analysis of proteins and RNAs in biological specimens.

Key Features:

• Automated Spectral Signature Extraction: Uses Vertex Component Analysis (VCA) to extract spectral signatures from multispectral fluorescence images.
• Remote Sensing Unmixing Algorithms: Incorporates three distinct unmixing algorithms adapted from remote sensing and validated on four biological datasets.
• Benchmarking: Performance has been compared and benchmarked against ZEN Black (Zeiss).
• Integration for smFISH Quantification: Integrated into computational pipelines that quantify individual RNA transcripts at single-cell resolution within intact tissues and support differential expression analysis of smFISH data.
• Handling of Challenging Samples: Addresses mixed signals and dye bleed-through in challenging specimens, including postmortem human brain tissue.

Scientific Applications:

• Multiplexed detection: Measure multiple proteins and RNAs simultaneously in a single imaging session in tissue slices or cultured cells.
• Single-cell transcript quantification: Quantify individual RNA transcripts at single-cell resolution in intact tissues using smFISH.
• Quantitative analysis of complex specimens: Enable accurate quantitative analyses in samples with mixed signals and dye bleed-through, such as postmortem human brain tissue.

Methodology:

Computational methods include Vertex Component Analysis (VCA) for automated spectral signature extraction, three remote-sensing-derived unmixing algorithms, benchmarking against ZEN Black (Zeiss), and integration into pipelines for single-cell smFISH transcript quantification and differential expression analysis.