This repository contains a set of image analysis scripts used in the study “Processing of DNA polymerase-blocking lesions during genome replication is spatially and temporally segregated from replication forks” by Wong et al. (in press at Molecular Cell; DOI: 10.1016/j.molcel.2019.09.015). In this study, all image analyses were automated using unbiased, customized scripts written in ImageJ macro and Jython languages with ImageJ FIJI software (https://fiji.sc/). The following analyses were used:

1. 2D foci counting (Related to Figures 1D, 1G, S1A and S1B)

   This script determines the number of foci of a fluorescent protein in individual nuclei. It analyzes images acquired with three channels, but uses only the two fluorescent channels for downstream analysis.

   • Input: Channel 1 (target fluorescent protein), Channel 2 (DAPI) and Channel 3 (bright field, not used)

   • Approach: The DAPI signal (Channel 2) is used to create a nuclear mask. Foci (Channel 1) are then segmented in 2D by auto-thresholding, and the number of foci per nucleus is determined for each nuclear mask.

   • Critical parameters to optimize: choice of auto-thresholding method(s) for nuclei and foci

   • Output: a table listing the number of foci for each nucleus analyzed

2. 3D foci counting (Related to Figures 2A, S2A, S2B and S2C)

   This script determines the number of foci of a fluorescent protein in individual nuclei. It performs foci segmentation in 3D to avoid fusion of foci in lateral proximity, but in different planes upon Z projection. It analyzes images acquired with three channels, but uses only the two fluorescent channels for downstream analysis.

   • Input: Channel 1 (target fluorescent protein), Channel 2 (DAPI) and Channel 3 (bright field, not used)

   • Approach: Nuclear masks are created in 2D from the DAPI signal (Channel 2). A “3D TopHat” filter is then applied to the images of the foci (Channel 1) before auto-thresholding, and the number of foci per nucleus is determined for each nuclear mask.

   • Critical parameters to optimize: filtering parameter for “3D TopHat” filter and choice of auto-thresholding method(s) for nuclei and foci

   • Output: a table listing the number of foci for each nucleus analyzed

3. 3D foci counting, colocalization, intensity and volume quantification (Related to Figures 2B, 3B-D, 3G, 4A-D, 5A, 5C-D, 6C, 6G-H, 6K, S2D-F, S3A-E, S3K, S4A-B, S5A-F, S6B, S6E, S6G)

   This script performs an object-based colocalization analysis for two target proteins imaged in two different channels. It determines the number of foci per nucleus for both targets, their intensities and volumes as well as the degree of overlap with the other target. The background fluorescence of one of the targets is used to create nuclear masks; if this is not possible, a fluorescent nuclear marker should be introduced by other means.

   • Input: Channel 1 (target fluorescent protein 1 with exclusively nuclear localization) and Channel 2 (target fluorescent protein 2)

   • Approach: Nuclear masks are created from the pan-nuclear background signal of target protein 1 (Channel 1), using filtering and auto-thresholding. An absolute threshold is applied to each of the targets in 3D for foci segmentation, and the number of foci per nucleus is determined for each nuclear mask. The co-localization value for each object (focus) is calculated by the relative volume (in percent) of each object that is covered by the other target. To quantify total foci intensity, the 3D foci masks generated from segmented foci are used to measure intensities in the original fluorescence images (expressed as total integrated densities in arbitrary units). Foci volumes are calculated from foci masks (expressed as µm³ for scaled images or number of voxels for unscaled images).

   • Critical parameters to optimize: minimum size and absolute threshold for foci segmentation of each target and choice of auto-thresholding method for nuclear segmentation

   • Output: four tables in total – one table per target protein listing the number of foci per nucleus and one table per target protein listing the properties of each focus (degree of overlap with the other target protein, intensity and volume)

4. Zoning assay (Related to Figure 5B)

   This macro counts foci and determines the location of each focus with respect to three nuclear zones of equal volume. It requires a fluorescent channel that marks the nuclear volume and a target fluorescent protein.

   • Input: Channel 1 (nuclear signal) and Channel 2 (target fluorescent protein)

   • Approach: 3D nuclear masks are generated by processing the nuclear signal (Channel 1) with a “Laplacian of Gaussian” filter and auto-thresholding. The resulting 3D objects are tested for their circularity. Nuclei that do not pass a threshold value are removed before ellipsoid fitting. The fitted ellipsoids are eroded into three concentric 3D zones of equal volume. Images of the target fluorescent protein (Channel 2) are segmented with an absolute threshold, and each object (focus) is assigned to one of the three nuclear zones based on the localization of its center.

   • Critical parameters to optimize: minimum size and absolute threshold for foci segmentation of the target fluorescent protein and choice of the auto-thresholding method for the nuclear signal

   • Output: three tables in total – one table listing the number of foci per nucleus, one table listing the desired and actual dimensions of zones (for quality control), and one table listing the assignment of all foci to the nuclear zones

5. Quantification of total nuclear intensity (Related to Figures 3E, S3F, S3J, S6D)

   This script performs quantification of total and mean nuclear fluorescent intensities. It requires a fluorescent channel for generating a nuclear mask (DAPI) and a target fluorescent protein.

   • Input: Channel 1 (DAPI) and Channel 2 (target fluorescent protein)

   • Approach: Nuclear masks are created from the DAPI signal (Channel 1) using filtering and auto-thresholding. The nuclear masks are then used to measure nuclear area (expressed as µm² for scaled image and number of pixels for unscaled image) and to quantify total intensity of the target fluorescent protein (Channel 2) (expressed as total integrated density with an arbitrary unit) and mean intensity (total intensity divided by area).

   • Critical parameters to optimize: choice of auto-thresholding method(s) for nuclei and foci

   • Output: a table listing total nuclear intensity, area and mean intensity per nucleus

6. Foci tracking (Related to Figures 1F, S1D-F)

   This script tracks foci in 2D over time with the plugin TrackMate.

   • Input: a list of time-lapse movies of a fluorescent protein in 2D from individual cells, manually cropped to ensure that there is no major shift of cells and the entire nucleus is comprised within the z stack

   • Approach: Foci in a time-lapse movie are identified by a “Laplacian of Gaussian” (LOG) detector with filtering and thresholding. Spots are connected in time with appropriate constraints. Lifetime (in seconds) and number of merging and splitting events are determined for each track. Coordinates (expressed as µm for scaled image and number of pixels for unscaled image) and mean intensity (expressed as an arbitrary unit) are quantified for each focus.

   • Critical parameters to optimize: parameters for LOG detector (expected radius of foci and threshold), constraints for linking and splitting events (based on distances and quality of foci between frames), and filter for overall track quality

   • Output: two tables for each movie – one table listing the properties of tracks (lifetime, number of merging and splitting events) and one table listing coordinates and mean intensities of all foci in each frame