Collective motion across species and time

This repository contains the data and code accompanying the paper:

Papadopoulou M., Fürtbauer I., O’Bryan L.R., Garnier S., Georgopoulou D.G., Bracken A.M., Christensen C., and King A.J. (2022) "Dynamics of collective motion across time and species". Phil. Trans. R. Soc. B 20220068. https://doi.org/10.1098/rstb.2022.0068

The code provided is open source, but we kindly ask you to cite the above paper if you make use of it.

Data

The data comprise analysed tragectories of four species: stickleback fish, homing pigeons, goats and chacma baboons. Specifically, they include:

1. col_motion_metrics: metrics of collective motion for each event of collective motion of all species. The columns include: the unique id of events with duration more than 15 seconds (event), the 9 metrics that composed the swarm space, the duration of each event in sampling steps (event_dur) and the duration in seconds (event_dur_s).
2. relative_positions: bearing angles (in rad) of all group members to their closest 4 neighbors (rank 1 to 4) during all events of collective motion of all species. Plotted in Figure 2.
3. trajectories_eg: example trajectories of events of collective motion of all species. Plotted in Figure 1.
4. swarm_space: the results of the swarm space analysis, specifically the PCA output (pca_res.Rdata), the coordinates of all events in the PC1-PC2 space (pcs_data.csv), and in the tSNE1-tSNE2 space (tsne_data.csv), with group size information included. Plotted in Figure 2.
5. all_events_durs.csv: the summary of events per species. Columns include: ev_count (number of events in a given day), dur (total duration of events in a day), cdur (cumulative duration of events over days), tot_dur (the total duration of events across all days per species). Plotted in Figure 1.
6. pol_speed_data.csv: smoothed polarization and speed of all groups/dates across species. Plotted in Figure 1.
7. changepoints_stats.csv: the results of our changepoint (segmented regression) and stationarity (kpss tests) analysis. Created by the across_time_analysis.R file. Includes columns: var (metric name), theta_idx (the point in the events timeseries that the changepoint is identified), theta_s (the changepoint in seconds), sl (the confidence level of the changepoint identification), sl_bool (whether the changepoint is significant), ci_l and ci_h (lower and upper confidence interval of changepoint), ci_ls and ci_hs (the confidence intervals in seconds), boots_sd (the standard deviation of the changepoint from the bootstrap analysis), kpss_pvalue (the p-value of the kpss test for stationarity), kpss_bool (whether a timeseries is identified as stationary or not). Plotted in Figure 3.

The raw data are collected for the previous studies:

• Sticklebacks:

Georgopoulou DG, King AJ, Brown RM, Fürtbauer I. 2022 Emergence and repeatability of leadership and coordinated motion in fish shoals. Behavioral Ecology 33, 47–54. https://doi.org/10.1093/beheco/arab108

• Pigeons:

Sankey DWE, Storms RF, Musters RJ, Russell WT, Hemelrijk CK, Portugal SJ. (2021). "Absence of “selfish herd” dynamics in bird flocks under threat". Current Biology. https://doi.org/10.1016/j.cub.2021.05.009

• Goats:

Sankey DWE, O’Bryan LR, Garnier S, Cowlishaw G, Hopkins P, Holton M, Fürtbauer I, King AJ. 2021 Consensus of travel direction is achieved by simple copying, not voting, in free-ranging goats. Royal Society Open Science 8, rsos.201128, 201128. https://doi.org/10.1098/rsos.201128

• Baboons:

Bracken AM, Christensen C, O’Riain MJ, Fürtbauer I, King AJ. 2022 Flexible group cohesion and coordination, but robust leader–follower roles, in a wild social primate using urban space. Proceedings of the Royal Society B: Biological Sciences 289, 20212141. https://doi.org/10.1098/rspb.2021.2141

Code

The code reproduces the main figures and across time analysis of the study. All analysis is performed in R, version 4.1. Files:

• Fig1.R: reproduces Figure 1.
• Fig2.R: reproduces Figure 2.
• Fig3.R: reproduces Figure 3.
• across_time_analysis.R: performs the timeseries analysis on the metrics of collective motion across species (changepoint analysis, bootstrap, and kpss stationarity test), producing the changepoints_stats.csv file, plotted in Figure 3.

Package dependencies: dplyr, tseries, lm.br, boot, stringr, reshape2, raster, ggplot2, cowplot, wesanderson, (extrafont)

Funding

This work is supported by an Office for Naval Research (ONR) Global Grant (N629092112030) awarded to Andrew J. King.

Contact

• For any further information, email Marina Papadopoulou at: m.papadopoulou.rug@gmail.com