FishFeats: streamlined quantification of multimodal labeling at the single-cell level in 3D tissues

Characterizing the distribution of biological marker expression at the single cell level in whole tissues requires diverse image analysis steps, such as segmentation of cells and nuclei, detection of RNA transcripts (or other staining), or their mapping (e.g. assigning nuclei/RNA dots to their corresponding cell). Several software programs or algorithms have been developed for each step independently but integrating them into a comprehensive pipeline for the quantification of individual cells from 3D imaging samples remains a significant challenge. We developed FishFeats, an open-source and flexible napari plugin, to perform all these steps together within the same framework, taking advantage of available and efficient software applications. The primary core of our pipeline is to propose a user-friendly tool for users who do not have a computational background. FishFeats streamlines extracting quantitative information from multimodal 3D fluorescent microscopy images (smFISH expression in individual cells, immunohistochemical staining, cell morphologies, cell classification) to a unified "cell-by-cell" table for downstream analysis, without requiring any coding. Our second focus is to propose and ease manual correction of each step to further improve accuracy, which can be critical for many biological studies.

FishFeats is open source under the BSD-3 license, freely available on github: https://github.com/gletort/FishFeats (DOI 10.5281/zenodo.17701225). FishFeats is developed in python, as a napari plugin for the user interface. Documentation is available in the github pages: https://gletort.github.io/FishFeats/. To report an issue using FishFeats or contributing to it please file an issue in the github repository https://github.com/gletort/FishFeats/issues.