Simulation Dataset: Air-breathing synchrony in juvenile Arapaima gigas reveals collective coordination under individual physiological constraints

This is a supplementary simulation dataset to reproduce Figs. 4-7 of the manuscript "Air-breathing synchrony in juvenile Arapaima gigas reveals collective coordination under individual physiological constraints" by Bartashevich et al.

General Structure

The archive is organized into four folders, each corresponding to a figure in the manuscript:

• Fig4/
• Fig5/
• Fig6/
• Fig7/

All simulation data are provided as NumPy .npy files, except for Figure 7, which uses .csv format.

______________________________________________________________________________________________

- Fig4

This folder contains data for homogeneous (Fig. 4A–F) and heterogeneous (Fig. 4G–K) group compositions.

File naming convention:
Files are named according to:

• Breathing type (pers1–pers6)
• Number of agents (e.g., Npop200)
• Coupling strength parameter (beta, ranging from 0 to 30)

File contents:

• histvalues: Raw breathing interval data from 10 simulation runs, each with 1000 time steps (dt = 0.01)
• mean: Mean breathing interval
• std: Standard deviation of breathing intervals
• laplace_emp_arr: Results of a two-sided Kolmogorov–Smirnov test against a Laplace distribution with empirical scale parameter b=3.55
• laplace_stst_arr: Results of a two-sided Kolmogorov–Smirnov test against a fitted Laplace-like distribution

Metric files contain arrays evaluated across all coupling strengths (beta = 0–30).

Note (heterogeneous groups in Fig. 4G-K):

• pers3: Groups composed of three breathing types
• pers6: Groups composed of six breathing types

______________________________________________________________________________________________

- Fig5

This folder contains all data required to reproduce Figure 5.

File naming follows the same conventions as in Fig4, with an additional parameter:

• perc_slow: Number of agents of the slowest breathing type (pers4)
  • Range: 0–67 for Npop100
  • Range: 0–133 for Npop200

______________________________________________________________________________________________

- Fig6

This folder contains data for simulations with partaking.

File naming convention includes:

• Number of agents (e.g., Npop200)
• Intra-cluster coupling strength (bintra: 5, 10, 15, 20, 25, 30)
• Inter-cluster coupling strength (binter: 0.0–1.0 in steps of 0.1)

File contents:

• sum_of_sqr_diff_arr: Data for Fig. 6A
• laplace_pvalues: Data for Fig. 6B
• std: Data for Fig. 6C
• histvalues, probs: Data for Fig. 6D–G

______________________________________________________________________________________________

- Fig7

This folder contains three .csv files corresponding to Fig. 7A–C.

The files include the following columns used for visualization:

• compromise_id (x-axis)
• breathing_gr_r_corresponding (y-axis)
• para_id (color hue)

Visualization:
Plots can be reproduced using the scatterplot function from the Seaborn library in Python.

______________________________________________________________________________________________