Supplementary Data and Analysis:

One nest, two niches? Distribution of alimentary resources between an inquiline termite and its host.

Description

Here are the supplementary data and the code for the statistical analysis of the publication [1] (2024) .

Authors: Johanne Timmermans, Matsvei Tsishyn, Nicolas Fontaine, Yves Roisin

Installation and Usage

• Install python version 3.9 or later.
• Install python packages numpy and matplotlib with pip (or pip3 depending on your system):

pip install numpy
pip install matplotlib

• From the root directory of this project, execute python scripts generate_plot.py and generate_stats.py with:

python3 generate_plot.py
python3 generate_stats.py

Content

• Folder ./data/: contains all initial data measurements in .csv files.
• Folder ./fig/: contains all generated figures by the project.
• Folder ./stats/: contains all generated statistics by the project in .csv files.
• Files generate_plot.py and generate_stats.py: are the main scripts to generate all figures and statistics.
• Folder ./src/: contains some python functions and classes thar are dependencies of the main scripts.
• Separator of all .csv files is ;.

Bootstrap method description

To estimate standard errors and confidence intervals for the mean or median durations by activity type, we employed a bootstrap approach, as detailed below.

• For a given activity type, let D represent the set of recorded activity durations, and let n denote the number of durations in D.
• We generated 50,000 resampled lists Di​ (for i=1,…,50,000) by randomly selecting n durations from D with replacement.
• The standard error for the mean or median was estimated as the standard deviation of the means or medians across the 50,000 resampled lists Di​.
• The 95% confidence interval was calculated as the range from the 2.5th to the 97.5th percentiles of the means or medians across the resampled lists.

References

[1] One nest, two niches? Distribution of alimentary resources between an inquiline termite and its host. Johanne Timmermans, Matsvei Tsishyn, Nicolas Fontaine and Yves Roisin.