Long exposure to extreme heat magnifies the decoupling between bacterial resistance and recovery.
Climate change is intensifying the duration of heat waves and threatening community stability. Using soil Pseudomonas communities, we investigated how different heat pulse durations impact resistance (immediate response) and recovery (post-stress response). We first assessed thermal performance traits across six species (16 strains) and found no trade-off between growth rate and heat resistance. We then assembled synthetic communities and exposed them to single heat pulses of 6, 12, 24, or 48 hours. We expected the fastest and most heat resistant species, P. putida, to dominate where present, but an intermediate grower, P. protegens, dominated all communities due to diffusible toxins and unexpected heat tolerance due to density-dependent growth. On average, each additional hour of heat increased extinction risk by 21.5% but fast growing communities were protected from extinction. The longest heat pulse duration led to sharp losses in diversity and productivity. We found that longer heat exposure magnified the decoupling between resistance and recovery phases, reducing community stability. These results reveal that growth rate and species interactions — not heat resistance alone — shape community fate during extreme events. Our findings highlight the need to consider nonlinear dynamics and trait-based interactions in predicting microbial responses to climate extremes.
This GitHub repository was created to facilitate the reproducibility of the scientific article listed above. The preprint manuscript is available on EcoEvoRxiv at this link: https://doi.org/10.32942/X2RS71
The complete protocols, data, and analysis for the manuscript can be found in this repository. The only data that is not found here is the flow cytometry data and FCS Express analysis because those files are too big. You can find those on FigShare at the following project link: https://figshare.com/projects/Longer_heat_pulses_disrupt_bacterial_communities_by_decoupling_resistance_from_recovery/246812
As detailed in the manuscript, the project is broken up into two experiments that are creatively called "Experiment I" and "Experiment II". This repository is laid out in the same way, with each experiment's main folder (prefixed "expt") containing dedicated subfolders for "protocols", "data", and any "old_files". The finalized code for the analysis and the manuscript figures are in the main experiment folders (along with any intermediate files outputted by the analysis scripts). The finalized code for the analysis is written in a notebook format (".Rmd" file extension) and then typeset (".html" file extension). The figures from the main text are outputted by these scripts in png format; the figures from the supplement can be found in the typeset notebooks.
The main folder for Experiment I is called "expt1_traits". The finalized analysis is found in the R Notebook called "thermal_performance_traits". Its typeset html file contains figures that are found in the supplement. As stated above, this main folder contains intermediate files outputted by the script: the png files for Figure 2 of the main text and intermediate data stored in an RData file. The "./data" subfolder contains txt files with the OD data for the growth curves, one xlsx file with the plate reader calibration data, and two xlsx files with the CFU data. The "./old_files" subfolder contains a more complete calibration of the microplate spectrophotometers (in the subsubfolder "./old_files/calibration") and a preliminary analysis of the growth rate estimates as well as Anjaney Pandey's final presentation at the end of his internship (in the subfolder "./old_files/TTD").
The main folder for Experiment II is called "expt2_cocultures". The finalized analyses are found in 2 R Notebooks called "main_expt--flow_cytometry_analysis" and "main_expt--OD_analysis". The raw data is found in the subfolder "./raw_data", intermediate data produced by the R Notebooks is in the subfolder "./intermediate_data", and png figures for Figures 3-5 of the main text are produced by the R Notebooks into the subfolder "./figures".
The 2 R Notebooks files with the prefix "main_expt--" are dependent on one another because they each create csv or RData files with intermediate data that is used by the other (e.g., indicating well annotation, extinction, and contamination). This is slightly annoying when running each of those scripts independently as you will need to run "main_expt--flow_cytometry_analysis.Rmd" first (it will run about 1/4 of the way before failing), then run "main_expt--OD_analysis.Rmd" (this will run completely without any issues), and finally you will be able to run "main_expt--flow_cytometry_analysis.Rmd" without any issues. But, if you simply download the entire git repository, you should be fine.
The subfolder "./raw_data" contains several subsubfolders that are prefixed with "serial_transf--" followed by a date (i.e., corresponding to the starting date of that experiment). This is the data for the serial transfer experiment: csv files with the flow cytometry cell counts from FCS Express, xlsx files with the flow cytometry well volume from Attune, and txt files with the OD data.
The raw data and analysis of the supernatant (i.e., spent media) experiments are also found in the "expt2_cocultures" folder. The R script is called "supernatant_expt_code.R" and its associated raw data is in the subfolder "./raw_data" in the file called "supernatant_expt_data.csv".
Finally, there is a main folder called "writing". Here you can find docx files for drafts of the main text and supplement, as well as image files where multipanel figures were combined into their final format for the main text. The main text was written and revised in a shared folder on Overleaf.
For Experiment I: the growth curve protocol was co-written by Anjaney Pandey (AP) & Ana-Hermina Ghenu (AHG), the growth curve data was gathered by AP as supervised by AHG, the growth curve data was analyzed by AHG with input from AP. The CFU protocol, data, and analysis was done by AHG.
For Experiment II: the serial transfer protocol and data was done by AHG with help from Anine Wyser, and then analyzed by AHG. The supernatant experiments were performed and analyzed by Zachary Michael Bailey (ZMB).