Three-dimensional scoring of zebrafish behavior unveils biological phenomena hidden by two-dimensional analyses
Abstract The study of zebrafish behavior represents a cornerstone upon which basic researchers promise to advance knowledge in life sciences. Although zebrafish swim in a three-dimensional (3D) space, their behavior in the lab is almost exclusively scored in two dimensions, whereby zebrafish are rec...
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2017
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/c16816d1527b4c0dabcae725e220b2a7 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Abstract The study of zebrafish behavior represents a cornerstone upon which basic researchers promise to advance knowledge in life sciences. Although zebrafish swim in a three-dimensional (3D) space, their behavior in the lab is almost exclusively scored in two dimensions, whereby zebrafish are recorded using a single camera providing 2D videos. Whether this dimensional reduction preserves the reliability of data has not been addressed. Here we show that, compared to a 3D observation, 2D data are flawed by over-reporting and under-reporting of locomotory differences. Specifically, we first reconstructed 3D trajectories through the integration of synchronous information derived from two cameras, and then compared them with the original 2D views in classical experimental paradigms assessing shoaling tendency, fear, anxiety, and general locomotion. Our results suggest that traditional behavioral scoring of individual zebrafish performed in 2D may undermine data integrity, thereby requiring a general reconsideration of scoring zebrafish behavior to incorporate a 3D approach. We then demonstrate that, compared to 2D, a 3D approach requires a reduced number of subjects to achieve the same degree of validity. We anticipate these findings to largely benefit animal welfare by reducing the number of experimental subjects, without affecting statistical power. |
|---|