Prey Size Analysis

This depository contains the code & datasets used in Santin et. al, Modulation of prey size reveals adaptability and robustness in the cell cycle of an intracellular predator., 2022. In this work, time lapse microscopy was used to monitor the Bdellovibrio cell cycle at the single cell level. The depository can be divided in two parts: the analysis pipeline used to obtain the data, and the experiments, where the datasets from different experiments are gathered. The experiments folder also contains the scripts used for the visualization of each dataset.

Pipeline

The pipeline is written as an R markdown file Pipeline_final.Rmd, where the user can choose to run the complete script or run the analysis chunk-by-chunk. The pipeline loads the miscroscopy & segmentation data (see below), detects the popping time, and (depending on the data type) detects & counts foci and/or performs growth analysis based on cytoplasmic fluorescence.

Data curation: before running the pipeline

The pipeline contains code to process movies of bdellovibrio containing fluorescent foci and/or cytoplasmic signal & uses phase-contrast to detect the bdelloplasts. As input it needs folders containing

• A .tiff or .tiffs of the full phase contrast movie(s); aligned so there is no drift (for instance using Image Stabalizer )
• A .tiff or .tiffs of the full fluorescent movie(s); aligned so there is no drift
• A .tiff or .tiffs of only the first frame of the phase movie(s) (For outline detection if you're using Oufti, see below)
• The output of either Oufti or MicrobeJ segmentation of the bdelloplasts in the first frame of the phase contrast movies. This pipeline was tested with Oufti.

Data curation: folder architecture

Sub-folders representing your conditions and replicates must be inserted in the phase-contrast and fluorescent folders according to the following architecture:

Phase-contrast/Fluorescence

• Condition A
  • Replicate 1 (put all your tiffs for Phase-contrast/Fluorescence for Condition A Replicate 1 here)
    • Image.tiff
  • Replicate 2 (put all your tiffs for Phase-contrast/Fluorescence for Condition A Replicate 2 here)
• Condition B
  • Replicate 1
  • Replicate 2 ...

Sub-folders representing your conditions must be inserted in the outline folder according to the following architecture (This is usefull if you need different Oufti parameter for each condition as you'll end with one Oufti file per batch) :

Phase-contrast/Fluorescence

• Condition A
  • outlines.mat (or outlines.csv if you're using microbeJ)

Software requirements

The pipeline uses the following software:

• R ≥ 4.2.2
• ImageJ2 with the Trackmate Plugin (standard in the FIJI distribution of ImageJ)
• BFconvert from BioFormats bftools. (Download & unzip bftools.zip here)

Other notes/requirements

• It is convenient but not required to run R chunks from Rstudio
• The pipeline was used & tested in Windows 10

The parameter file

This file parameters.yml contains all the settings the pipeline needs to know to run properly. This is changed for each experiment. You can find a parameters.yml file in each Experiments subfolder. This file contains the paths to the required softwares & datasets and the parameters used to compute the output data. When working with new data, make sure to at least change the folder paths & check the Microscopy settings (pixel to micron conversion & minutes per frame).

Experiments

The experiments folder contains the data output & visualization code used in Santin et al. This is divided in experiments exploring the Bdellovibrio growth rate (folder Growth_rates) and the number of Progenies and Origins of Replication (folder Progenies_&_cell_cycle).

Parameters, output files obtained from our pipeline and .rmd file to reproduce the plots are organized as follows:

• Pipeline
• Experiments
  • Growth rates
    • LB E. coli
    • M9 E. coli
  • Progenies & cell cycle
    • LB E. coli
    • M9 E. coli
    • LB Citrobacter
      • Parameters.yml
      • original pipeline outputs.rmd
      • compiled and summarized data.csv
      • Plots.Rmd

Issues & requests

If you encounter errors, have a question or a request, please open an Issue or send an email to the Laloux lab.

Authors

This pipeline is created & maintained by Thomas Lamot (main) and Renske van Raaphorst

Cite

Y.G. Santin, T. Lamot, J. Kaljevic, R. van Raaphorst, G. Laloux (2022), Modulation of prey size reveals adaptability and robustness in the cell cycle of an intracellular predator., in submission