Creation of High‐Dimensional Reduction Analysis‐Compatible Histocytometry Files from Images of Densely‐Packed Cells and/or Variable Stain Intensity. Issue 5 (24th May 2022)

• Record Type: Journal Article
• Authors: Munoz‐Erazo, Luis; Schmidt, Alfonso J.; Shinko, Diana; Eccles, David A.; Price, Kylie M.
• Is Part Of: Current protocols. Volume 2:Issue 5(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-05-24
• Subjects: cell segmentation -- dimensional reduction analysis -- heterogenous -- histocytometry -- immunophenotyping -- microscopy; Life sciences -- Laboratory manuals -- Periodicals; Biology -- Laboratory manuals -- Periodicals; Life sciences -- Technique -- Periodicals; Biology -- Technique -- Periodicals; 570.028
• DOI: 10.1002/cpz1.441
• Languages: English
• ISSNs: 2691-1299

Abstract

Abstract: The power of high‐dimensional reduction techniques using multiparameter images has been demonstrated across a variety of different publications. Recently, we published an end‐to‐end low‐cost GUI‐based protocol for performing histocytometric spatial analysis on images derived from the most common microscope image formats. However, this protocol is limited by the normalized marker intensity outputs and the difficulty in processing images of highly aggregated and/or exceptionally heterogenous cell populations. Here we present the basic protocols required to construct an advanced histocytometric data file using only freeware. This data file is compatible with images containing cell nuclei clusters that are difficult to segment, and results in histocytometry files retaining the original marker intensity values of the microscopic images they were derived from. This is especially useful in cells that are phenotyped based on relative marker expression levels. Histocytometry data files produced by these protocols are compatible with high‐dimensional reduction analysis using marker intensity data, such as tSNEs. This methodology is showcased using stitched microscopic images of murine lymph nodes, complex organs with highly aggregated heterogenous cell populations, that are typically difficult to segment. © 2022 Wiley Periodicals LLC. Basic Protocol 1 : Image preprocessing and generation of nuclei marker probability maps Basic Protocol 2 : Cell segmentation using ilastik‐derived probability maps Basic Protocol 3 : Generation of histocytometric .fcs files