CeleST: Computer Vision Software for Quantitative Analysis of C. elegans Swim Behavior Reveals Novel Features of Locomotion
Figure 1
Summary of CeleST components and usage.
Input files are videos of multiple swimming C. elegans. Files are stored in a database that records identifying features (strain, date, etc.) to permit easy selection of animals to be compared by analysis. After selection of animals to be compared, swimmers are automatically tracked from videos, and computation from curvature data or posture is used to score ten swim measures in 30 second swim trials (see description in text). Measures from the scored animals are compiled and can be exported in several alternative data analysis formats, including dot plots, line graphs, histograms and two dimensional comparison (ellipses indicate the principal directions and the standard deviations of the data). Statistical analysis is automated. A dynamic demonstration of CeleST tracking and computing of measures can be found in Video S1 1–4 on http://celest.mbb.rutgers.edu.