zFISHer: Automated 3D Registration, Detection, and Colocalization with Interactive Curation for Sequential...

Area of Science:

• Biotechnology
• Microscopy
• Bioinformatics

Background:

• Analyzing three-dimensional (3D) datasets from sequential multiplexed fluorescence in situ hybridization (FISH) for puncta colocalization is a labor-intensive bottleneck.
• Spatially resolved molecular profiling in cell monolayers using FISH requires efficient 3D data analysis.

Purpose of the Study:

• To develop an open-source application, zFISHer, for automated and interactive analysis of sequential FISH image data.
• To overcome the limitations of manual puncta colocalization analysis in 3D FISH datasets.

Main Methods:

• zFISHer utilizes the napari viewer for an end-to-end analysis pipeline.
• Includes automated nuclear segmentation, puncta detection on unaligned z-stacks, and multi-round image registration with optional B-spline deformable warping.
• Features a "Fishing Hook" raycasting algorithm for precise 3D puncta centroid localization and sub-voxel volume optimization.

Main Results:

• zFISHer provides complete automation of sequential FISH image processing with interactive user-curation tools.
• Achieves precise transformation of puncta coordinates into aligned space and consensus nuclei generation.
• Enables pairwise and tri-channel colocalization analysis with statistics and high-throughput batch processing.

Conclusions:

• zFISHer offers a comprehensive solution for analyzing 3D FISH data, significantly reducing analysis time and improving accuracy.
• The application's automated and interactive features facilitate efficient molecular profiling and colocalization studies.
• zFISHer is freely available as open-source software, promoting wider adoption in biological research.