Multiview motion tracking based on a cartesian robot to monitor Caenorhabditis elegans in standard Petri dishes

Abstract

Data from manual healthspan assays of the nematode Caenorhabditis elegans (C. elegans) can be complex to quantify. The first attempts to quantify motor performance were done manually, using the so-called thrashing or body bends assay. Some laboratories have automated these approaches using methods that help substantially to quantify these characteristic movements in small well plates. Even so, it is sometimes difficult to find differences in motor behaviour between strains, and/or between treated vs untreated worms. For this reason, we present here a new automated method that increases the resolution flexibility, in order to capture more movement details in large standard Petri dishes, in such way that those movements are less restricted. This method is based on a Cartesian robot, which enables high-resolution images capture in standard Petri dishes. Several cameras mounted strategically on the robot and working with different fields of view, capture the required C. elegans visual information. We have performed a locomotion-based healthspan experiment with several mutant strains, and we have been able to detect statistically significant differences between two strains that show very similar movement patterns.

Figure 1

Multi-view cartesian robot. The image was obtained with SOLIDWORKS 2020 SP3.0. This is the general view, showing all components.

Figure 2

The XY carriage. The image was obtained with SOLIDWORKS 2020 SP3.0. (a) The whole carriage, (b) a section view to see parts in detail.

Figure 3

The head. The image was obtained with SOLIDWORKS 2020 SP3.0. (a) The head has two microcameras and one laser. (b) The head and de carriage are assembled, where the laser light trajectory is shown.

Figure 4

ROS diagram. Black squares are ROS nodes running on desktop computer, green squares are the four Raspberries Pi and blue squares are ROS nodes running on Raspberries Pi. Macroprocessing_Node implements the macroimage processing, macrotracking and state machine. Microimage_Node implements both microimage processing, microtracking and microimage focus (by changing Z axe). The nodes: cam_1_Micro_Node, cam_2_Micro_Node and cam_1_Macro_Node, execute the camera driver, take the image and publish the images into /microImage_1, /microImage_2 and /macroImage_1 respectively.

Figure 5

State machine. This flowchart shows the tasks of the state machine. Macrotracking is running constantly (it is software tracking), while state machine executes (1) when the worm has been selected. The 2-state starts when worm-goal has been achieved (the worm selected is framed). And when a 30-s sequence has been captured, then the mechanical tracking ends and the loop restarts.

Figure 6

Module index. Every worm movement of every time instant is recorded (worm 1, worm 2 ... worm n).

Figure 7

The module index for automated inspection and the thrashing index for manual inspection are shown, which are both transformed into percentages to facilitate visual comparison. On the left is the module sum index. On right, is the graphical output of the manual analysis of the movement capacity. Strains for both experiments are wild type and mutant worms. Wild type (N2 Bristol), unc-1 mutants and transgenic worms expressing CAG expansions in GABAergic neurons show very different motor capacity in liquid. **Means $p\text{-value}<0.05$, with Wilcoxon rank sum test. ***Means $p\text{-value}<0.001$, in ANOVA test with Tukey’s post-hoc test. The image was drawn with MATLAB R2020b by using the ’boxplot’ function.

Figure 8

The head and the carriage assembly are shown and how all components interact. The image was obtained with SOLIDWORKS 2020 SP3.0. (a) Shows a microimage taken from one of both microcamera. (b) Shows macroimage taken from macrocamera.

Figure 9

Relative frequency of module values. The abscissa axis represents the module value rank. The units are in steps, which in millimetres are $1.09\mathrm{\mu }\mathrm{m}/step$. The blue bars are Wild Type (N2) strain, the orange bar is vltIs66[unc-25p::144CAG] strains and the red one is unc-1(vlt10) strain. The plot was drawn with the ’bar’ function of MATLAB R2020b.