Supplementary data for the vassi Python package

We provide intermediate results (examples/social_cichlids/results.h5 and examples/social_cichlids/k_fold_predictions.h5) for the examples/social_cichlids/results_and_figures-cichlids.ipynb notebook (available in GitHub repository or vassi_source.zip. This notebook reproduces the figures as presented in our paper. We also provide the results obtained from hyperparameter optimization using the optuna framework in the same directory (examples/social_cichlids/optimization/*).

The results from k-fold prediction on the entire dataset (for visualization of networks as presented in the paper) are available in examples/social_cichlids/k_fold_predictions_predictions.csv, which can be loaded as a dataset when complemented with the trajectories file (see vassi_source.zip/examples/social_cichlids/results_and_figures-cichlids.ipynb for details). Our paper also presents a comparison between model predictions (behavior counts) and association time as an alternate behavioral proxy for interactions.

The raw data files and the corresponding r script are available at examples/social_cichlids/predictions_vs_association. 2. CALMS21 dataset In addition, we tested our package on an existing benchmark dataset, the CALMS21 mouse resident-intruder dataset. For convenience, we provide Python scripts to download or convert the original dataset (vassi_source.zip/src/vassi/case_studies/calms21/download.py and vassi_source.zip/src/vassi/case_studies/calms21/convert.py; or available after vassi was installed, see online documentation for more details).

The original CALMS21 dataset can be downloaded here: data.caltech.edu/records/s0vdx-0k302 [Dataset] Jennifer J. Sun, Tomomi Karigo, David J. Anderson, Pietro Perona, Yisong Yue, & Ann Kennedy. (2021). Caltech Mouse Social Interactions (CalMS21) Dataset (1.0) [Data set].

CaltechDATA. doi.org/10.22002/D1.1991 [Paper] Sun JJ, Karigo T, Chakraborty D, Mohanty SP, Wild B, Sun Q, Chen C, Anderson DJ, Perona P, Yue Y, Kennedy A. The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactions. Adv Neural Inf Process Syst. 2021 Dec;2021(DB1):1-15. PMID: 38706835; PMCID: PMC11067713.

All scripts and notebooks from which results were presented in the paper used the same configuration for feature extraction (examples/CALMS21/features-mice.yaml). As for the other example dataset, we provide intermediate results (examples/CALMS21/results.h5) for the vassi_source.zip/examples/CALMS21/results_and_figures-mice.ipynb notebook to reproduce the figures presented in our paper. We also provide the results obtained from hyperparameter optimization using the optuna framework in the same directory (examples/CALMS21/optimization/*).

Files datasets/ └── social_cichlids/ ├── videos/ │ ├── GH010423.MP4 │ ├── GH010861.MP4 │ ├── GH013974.MP4 │ ├── GH019910.MP4 │ ├── GH030423.MP4 │ ├── GH030451.MP4 │ ├── GH030861.MP4 │ ├── GH039910.MP4 │ └── GH039931.MP4 ├── cichlids_annotations.csv └── cichlids_trajectories.h5 examples/ └── CALMS21/ ├── optimization/ │ ├── optimization-results.yaml │ ├── optimization-summary.yaml │ └── optimization-trials.csv ├── features-mice.yaml └── results.h5 └── social_cichlids/ ├── optimization/ │ ├── optimization-results.yaml │ ├── optimization-summary.yaml │ └── optimization-trials.csv ├── predictions_vs_association/ │ ├── aggregated_counts-1bl.csv │ ├── aggregated_counts-3bl.csv │ ├── aggregated_counts-5bl.csv │ └── predictions_vs_association.Rmd ├── features-mice.yaml ├── results.h5 └── k_fold_predictions_predictions.csv └── vassi_source.zip