Fig. S1

Measurement of embryonic motor activity. (A) Schematic of the apparatus used to record motor activity in embryos. Embryos are arrayed in a 49 well grid in a Petri dish with temperature maintained at 28°C using a thermistor and heating element. A camera mounted above records images (B) at a 1 Hz framerate. (C) Motor activity is measured as the sum of the number of pixels which change in intensity greater or less than a predetermined threshold between frames. Top row shows a sequence of frames for the boxed well in (B), while the bottom row shows the same frames, with changed pixels highlighted and the total number of changed pixels indicated. To allow real-time processing, only every second pixel in the image is checked. (D) For validation, the coiling frequency was manually observed for 20 embryos (number of coiling events in a one minute interval) and the total number of pixel change events automatically measured for the same video sequence. Manual and automatically scored values are highly correlated Pearson r=0.86, P<0.001 confirming that the algorithm captures differences in motor activity. (E) For additional validation, embryos were exposed to varying concentrations of tricaine for 20 minutes (N=14 each condition) before movement was measured over a 10 minute period. The y-axis shows the mean number of pixel changes per minute averaged over the recording period. 0.002% tricaine significantly reduced but did not eliminate movement while 0.02% tricaine completely paralyzed embryos such that the number of pixel changes was not significantly (n/s) different from baseline as measured in wells with no embryos. * P<0.001.