TumbleScore: Run and tumble analysis for low frame-rate motility videos

Abstract

Scientists often exploit the motility of peritrichously flagellated bacteria for various applications. A common alteration is modifying the frequency of mid-movement changes in direction, known as tumbles. Such differences in bacterial swimming patterns can prove difficult to quantify, especially for those without access to high-speed optical equipment. Traditionally, scientists have resorted to less accurate techniques, such as soft agar plate assays, or have been forced to invest in costly equipment. Here, we present TumbleScore, software designed to track and quantify bacterial movies with slow, as well as fast, frame-rates. Developed and fully contained within MATLAB, TumbleScore processes motility videos and returns pertinent tumbling metrics, including: (i) linear speed, (ii) rotational speed, (iii) percentage of angle changes below a given threshold, and (iv) ratio of total path length to Euclidian distance, or arc-chord ratio (ACR). In addition, TumbleScore produces a "rose graph" visualization of bacterial paths. The software was validated using both fabricated and experimental motility videos.