Computational protocol: Pre and Post-copulatory Selection Favor Similar Genital Phenotypes in the Male Broad Horned Beetle

Similar protocols

Protocol publication

[…] After killing by freezing, experimental male beetles were placed on a slide, oriented consistently and digital images of the thorax were captured using a Leica M125 microscope with a mounted digital camera (Leica DFC 295) that was linked to a PC. Next, the aedeagus was removed from the abdominal cavity, placed on a glass slide so that the anterior tip of the aedeagus was laterally oriented to the left and then the whole structure was mounted in a droplet of Hoyer’s solution.The width of the pronotum was measured as an index of body size (see ) using Image J (version 1.48). Variation in the size and shape of the male genitalia was quantified using a geometric morphometric approach. First, three points that could be located precisely across all specimens were defined as fixed landmarks (type-two landmarks) and another 26 points were defined as semilandmarks as they slide along the curved outline of the aedeagus (). The fixed and semilandmarks were manually applied to all images in the same sequence (i.e., 1–29) in the program TPSDig 2.14 and the semilandmarks were identified by use of a “slider file” in the program TPSUTIL 1.46 (). Second, the landmark data were extracted in the program tpsRELW 1.46 () and normalized for position, orientation, and scale (i.e., Procrustes superimposition) to eliminate non-shape variation (Zelditch et al. 2012). Next, centroid size (i.e., the square root of the sum of squared distances between landmarks from the central point of the specimen) and relative warp (RW) scores were calculated in tpsRELW 1.46. Finally, changes in the shape of the aedeagus were visualized as shape deformations of thin-plate spline plots in tpsRELW 1.46. The analysis was conducted on the complete data set (i.e., mating success and insemination success data combined), so that centroid size and the RWs were in the same geometric space to allow direction comparison of the direction and form of selection in two contexts. The 29 landmarks that defined the outline of the aedeagus yielded 54 RWs. Each RW explained diminishing amounts of variation in shape, so we only interpret RWs 1, 2, and 3 as cumulatively, they explain greater than 80% of the variance in shape (). It is possible that we have overlooked important variation in genital shape by limiting our analysis to that which is described by RW1, 2, and 3. However, the amount of variation that is described by subsequent RWs is small and difficult to describe as the shape changes are subtle. Furthermore, the interpretation of the strength and form of selection on traits can be troublesome as the number of traits increase and therefore the number of nonlinear terms increase (). Additionally, from a biological perspective, genital differences across taxa and populations are usually not subtle, in fact they are used to describe taxonomic differences when general morphology is identical. So using RWs that capture most of the variation seems the best approach from that perspective also. Fig. 1We measured the repeatability of male body size and the repeatability of digitization of two images of the same male aedeagus using the R code provided in . The repeatability of body size and genital size and shape was high (Pronotum width = 0.989, 95% CIs = 0.985, 0.991; Genital size = 0.989, 95% CIs = 0.980, 0.998; RW1 = 0.936, 95% CIs = 0.887, 0.985; RW2 = 0.885, 95% CIs = 0.799, 0.971; RW3 = 0.795, CIs = 0.649, 0.941). […]

Pipeline specifications

Software tools TpsDIG, TpsUtil, TpsRelw
Application Macroscope & basic digital camera imaging
Organisms Homo sapiens, Ilex paraguariensis