Data analysis was carried out using custom scripts in MATLAB (MathWorks). To determine the motor rotational speeds accurately, the elliptical bead trajectory was transformed to a circle based on an ellipse fit, and the speed was calculated as described previously (22). The motor speed was filtered with a 10-point running average, resulting in a time resolution of about 0.03 s. The speed traces of the latex beads showed stepwise changes. Determination of the length and position of each segment of speed step was carried out using a step-finding algorithm described previously (13, 20). The number of stators in each segment was determined by dividing the mean value of speed in this segment by the mean speed increment per stator for the whole 20-min trace. As the step sizes of speed increment per stator were fairly constant for each motor, but with variations among different motors, the mean speed increment per stator was determined individually for each motor. The two key parameters in the step-finding algorithm are the minimum step size in speed (vm) and the minimum length of the dwell time (tm). As the average step size was 5.6 ± 0.6 Hz (fig. S1), we chose the minimum step size vm as 3 Hz. The minimum length of the dwell time (tm) was chosen to be 1 s. To validate the step-finding algorithm, simulating traces were generated as follows: Stochastic simulation of stator coming on and off the motor was performed starting with N = 5, with the On intervals following a double-exponential distribution with decay rates of 0.246 and 0.0082 s−1, and the Off intervals following a single-exponential distribution with a decay rate of 0.0111 s−1 (fig. S4). The speed change per stator was generated following a Gaussian distribution with mean and SD of 5.6 and 0.6 Hz, respectively, according to the experimental measurements. After generating the motor speed trace, a Gaussian-distributed noise with zero mean and SD of 3.25 Hz (obtained from experimental traces) was added to the simulation trace. The step-finding algorithm (vm = 3 Hz, tm = 1 s) was then applied to the simulated traces, an example of which is shown in fig. S9. The step-finding algorithm correctly identified all the speed segments with duration longer than 1 s (with an error probability below 0.2%), and the resulting interval distributions reproduced the double-exponential shape with consistent values of the decay rates. Changing the value of vm in the range of 2 to 4 Hz, and the value of tm in the range of 0.1 to 1.5 s, did not change the conclusions (fig. S10).

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.