Spatial image analysis of germ nuclei fluorescent intensity in C. elegans gonads

Overview

This repository contains QuPath v0.5.1 and R scripts designed to analyze H3K4me3 signal intensity in immunostained C. elegans gonads.

Imaging procedure

• Confocal imaging: Samples are imaged using 40X tile-scanned Z-stacks at a resolution of 2048x2048 pixels. The imaging depth is set to 3 μm to minimize nuclear overlap.
• Standardized settings: Laser settings remain constant across all samples to ensure uniform intensity quantification.
• Image processing: Z-stacks are converted into average intensity projections using Leica LAS-X software before further analysis.

QuPath cell detection

• Nuclei segmentation: Nuclei are segmented using StarDist2D in QuPath v0.5.1, applying the dsb2018_heavy_augment.pb model with predefined parameters.
• Channel linking: DAPI and H3K4me3 fluorescence channels are linked to unique cell identifiers to maintain consistency in analysis.
• Intensity quantification: H3K4me3 and DAPI intensities are quantified using the formula: integrated intensity = mean intensity * area.
• Normalization: The H3K4me3 intensity is normalized relative to DAPI signal and the maximum H3K4me3 intensity observed per gonad.

Cell projection onto the gonad axis

• Linearized representation: Each nucleus is mapped to the nearest segment of the gonad skeleton to create a linearized representation of the gonad.

Statistical analysis

• Spearman correlation: The relationship between cell position along the gonad axis and normalized H3K4me3 intensity is assessed using Spearman correlations, which test for positive non-parametric associations.

Workflow

1. In QuPath:

   • Open /QuPath/project.qpproj.
   • Upload raw tiffs. Each sample should have three confocal images: a T-PMT, a DAPI, and a H3K4me3 channel image.
   • On all T-PMT channel images, annotate a closed gonad shape and a gonad skeleton polyline. The gonad skeleton polyline must start at the distal tip (i.e., first point clicked is distal tip), otherwise the remaining analyses will not be interpretable.
   • Run /QuPath/scripts/main.groovy.

2. In R:

   • Run /scripts/analysis.R.

Citations

Bankhead P, Loughrey MB, Fernández JA, et al. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017;7(1):16878. DOI: 10.1038/s41598-017-17204-5

Schmidt U, Weigert M, Broaddus C, Myers G. Cell detection with Star-Convex Polygons. In: Frangi AF, Schnabel JA, Davatzikos C, Alberola-López C, Fichtinger G, eds. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018. Springer International Publishing; 2018:265-273. DOI: 10.1007/978-3-030-00934-2_30

R: A Language and Environment for Statistical Computing. https://www.R-project.org/


Gonad linearization methods inspired by:

Toraason E, Adler VL, Kurhanewicz NA, et al. Automated and customizable quantitative image analysis of whole Caenorhabditis elegans germlines. Genetics. 2021;217(3):iyab010. DOI: 10.1093/genetics/iyab010