Skip to content

README.md

Latest commit

 

History

History
172 lines (133 loc) · 7.4 KB

README.md

File metadata and controls

172 lines (133 loc) · 7.4 KB
Raw

Shaking regime and inoculum OD in gut-bacterial growth experiments

Analysis code for the study of how the hydrodynamic (shaking) regime of a microplate reader and the optical-density-to-cell-number relationship bias growth measurements of human gut bacteria — two routine, usually-ignored settings that turn out to act differently on different species.

Seven gut species — Bacteroides thetaiotaomicron (BT), Escherichia coli LF82 (EC), Faecalibacterium duncaniae (FD), Mediterraneibacter gnavus (MG), Roseburia intestinalis (RI), Segatella copri (SC) and Blautia hydrogenotrophica (BH) — were grown under three regimes (static, pulsed, continuous shaking) in strictly anaerobic 96-well plates, with paired flow-cytometry, OD-to-cell calibration, and microscopy sub-studies.

This repository contains the analysis code needed to reproduce the analyses and figures from the raw data.

Paper & data

  • Paper: Adedamola Godwin Daodu, Pallabita Saha, Gabriela Bravo-Ruiseco, Xingjian Zhou, Karoline Faust. (Journal and DOI to be added on acceptance.)
  • Data: plate-reader, calibration, flow-cytometry and microscopy data are openly available on Zenodo: https://doi.org/10.5281/zenodo.23215813
  • Code archive: each release of this repository is archived on Zenodo; the DOI of the version used in the paper is given in the paper's Availability of data and materials statement.

The data are not stored in this repository. Download them from the Zenodo record above.

What's here

15 R scripts, one stage per file, run in numeric order. Scripts 01–05 do the analysis and write intermediate result tables; 06–12 produce the figures.

Script Purpose
00_load_and_filter.R Load the annotated data tables, keep included wells, check the design
00_run_all.R Run scripts 01–05 in order, with checks and logging
01_design_and_identifiability.R What the design can and cannot estimate (regime vs batch)
02_endpoint_OD.R Within-run endpoint-OD contrasts per regime
03_PvsC_paired.R Pulsed-vs-continuous analysis and flow cytometry
04_batch_effects.R Between-plate (run) variance components
05_CoV_reproducibility.R Within-plate coefficient-of-variation reproducibility
05_FD_growthcurve_startingOD.R F. duncaniae growth curves at two starting ODs
06_figures.R Main figures
07_supplementary_kinetics.R Supplementary kinetic figures
08_calibration_fits.R Per-species OD-to-cell calibration fits
09_calibration_figures.R Calibration figure
10_endpoint_conversion_diagnostic.R Endpoint calibration-extrapolation diagnostic
11_microscopy_scoring.R Blinded morphology scoring and rater agreement
12_bactoscoop_size.R Single-cell size extraction

A data dictionary (every column of every data file) accompanies the data on the Zenodo record.

Data files read by the scripts

File Contents
MASTER_kinetic_annotated.csv Shaken plates (pulsed, continuous): per-well endpoint OD and kinetic metrics, with inclusion/exclusion annotation
MASTER_static_annotated.csv Static plates: per-well endpoint OD, with inclusion/exclusion annotation
MASTER_timeseries_annotated.csv Shaken plates: per-well OD time series
SC_RI_flow_cytometry_master.xlsx Endpoint flow-cytometry counts for S. copri and R. intestinalis
calibration_metadata.csv OD-to-cell-count calibration (target OD, cells/mL per well)
microscopy_unblinded_scores.csv Blinded morphology scores from both raters, with the unblinding key
microscopy_size_results_by_slide.csv Per-slide median cell width, length and aspect ratio (E. coli, B. thetaiotaomicron)

Script outputs and manuscript figures

Output file Manuscript Script
Fig2_calibration Fig. 2 09
Fig3_variation Fig. 3 06
Fig4_endpoint_OD Fig. 4 06
Fig5_flow_cytometry Fig. 5 06
Fig6_morphology Fig. 6 11
FigS1_calibration_fits Fig. S1 09
FigS2_spread_vs_growthrate Fig. S2 06
FigS3_growth_curves, FigS4_kinetic_metrics, FigS5_kinetic_CoV Figs S3–S5 07
FigS6_cell_size Fig. S6 12
FigS7_predicted_vs_measured Fig. S7 10

Fig. 1 is a schematic and is not generated by code.

Requirements

  • R ≥ 4.5.0
  • R packages: tidyverse, lme4, lmerTest, emmeans, irrCAC, patchwork, ggrepel, broom, readxl

Install the packages:

install.packages(c("tidyverse","lme4","lmerTest","emmeans","irrCAC",
                   "patchwork","ggrepel","broom","readxl"))

Versions used for the published analysis: R 4.5.0; tidyverse 2.0.0, dplyr 1.1.4, ggplot2 4.0.0, readr 2.1.5, readxl 1.5.0, lme4 2.0.1, lmerTest 3.2.1, emmeans 2.0.4, irrCAC 1.4, patchwork 1.3.2, ggrepel 0.9.8, broom 1.0.10.

Reproduce it

  1. Get the code — clone or download this repository: 
    git clone https://github.com/msysbio/shaking-regime-gut-bacteria.git
  2. Get the data — download the data files from the Zenodo record (DOI above). Download them and place them in the same folder as the scripts: the scripts and data must sit together in one flat directory.
  3. Set the folder path at the top of a script (or your session): 
    DATA_DIR <- "/path/to/this/folder"
  4. Run the core analysis (scripts 01–05, in order): 
    source(file.path(DATA_DIR, "00_run_all.R"))
  5. Generate the figures (run as needed): 
    source(file.path(DATA_DIR, "06_figures.R"))                # main figures
source(file.path(DATA_DIR, "07_supplementary_kinetics.R")) # supplementary kinetics
source(file.path(DATA_DIR, "08_calibration_fits.R"))       # calibration fits
source(file.path(DATA_DIR, "09_calibration_figures.R"))    # calibration figure
source(file.path(DATA_DIR, "10_endpoint_conversion_diagnostic.R"))
source(file.path(DATA_DIR, "11_microscopy_scoring.R"))
source(file.path(DATA_DIR, "12_bactoscoop_size.R"))
    (Run 08 before 09 and 10, which use its output.)

All scripts read and write to DATA_DIR; figures are written there as PNG and PDF.

Notes

  • Flat folder. Scripts assume code and data share one directory; moving files into subfolders requires editing the DATA_DIR paths.
  • Statistics. Endpoint OD is modelled per species as endpoint ~ regime + (1|run) (REML; lme4/lmerTest), with pairwise contrasts via emmeans (Bonferroni-corrected within species) and rater agreement as Gwet's AC2 (irrCAC). Pulsed and continuous shaking are never compared directly — they never share a run, so each regime is compared only against its paired same-run static control.
  • Reproducibility. 00_run_all.R runs each script in a clean environment; scripts communicate only through files on disk.

Citation

If you use this code or data, please cite the paper:

Adedamola Godwin Daodu, Pallabita Saha, Gabriela Bravo-Ruiseco, Xingjian Zhou, Karoline Faust. (Title, journal, and DOI to be added on acceptance.)

and the Zenodo data record (DOI above) and code archive (DOI in the paper).

License

  • Code: CC BY-NC 4.0 (see LICENSE) — free to use, adapt and share with attribution; not for commercial use or sale
  • Data (on Zenodo): CC BY 4.0

Contact

Adedamola Godwin Daodu — Laboratory of Molecular Bacteriology, KU Leuven. Adedamolagodwin.daodu@kuleuven.be