Picture from Crombie, Timothy A et al. (2022) MicroPubl Biol

Figure 1. An automated assay workflow yields robust chemotaxis measurements.(A) The experimental workflow is shown. (A1-3) Nematodes are grown for three generations prior to being transferred to chemotaxis assay plates as synchronized fourth larval stage animals (L4s). (A4) Compounds are added to the chemotaxis assay plates using the guides (gray) that are laser etched into the custom acrylic plate holder and are visible through the NGMA plates. (A5) The animals are left to respond to cues for one hour at 20°C then the plates are transferred to 4°C before imaging. (A6) The image processing steps performed by ct are shown. The steps result in an output file with the chemotaxis index (CI) calculated for each image passed to the ct command. (B) A map of isolation locations for the 11 strains used in this experiment (red points), the other wild C. elegans strains available from the Caenorhabditis elegans Natural Diversity Resource (CeNDR) are also shown (gray points) (release 20210121) (Cook et al. 2017). (C) The genetic relatedness of CeNDR strains are shown as an unrooted tree, and the 11 strains used here are colored red and the other 529 CeNDR strains are colored black. (D) The relationship between manual CI calculations and automated CI calculations for identical images is shown. The coefficient ofdetermination (R2) and the linear regression equation are shown for each compound. The linear models were calculated using CIs of all 11 strains for either 1-octanol or isoamyl alcohol. (E) To test for a strain-dependent systematic error in our method, we fit a regression model of the form (automated CI ~ manual CI + strain). The strain coefficient estimates from the regression model are shown normalized to the reference strain, N2. The points represent the normalized estimate values and the horizontal lines represent 95% confidence intervals around the estimates. We did not detect a systematic effect of strain on method agreement by ANOVA (F(10,181) = 1.51, p = 0.138).