Classifying camel breeds using geometric morphometrics: A case study in Kuwait

Abstract

In this study, we examine the utility of image-based geometric morphometrics in classifying camel breeds. As a case study, we explore morphometric variation of six common Arabian Peninsula camel breeds (“Mezayen” breeds), traditionally named according to their coat color: Homor (red), Majaheem (black), Shaele (brown), Shageh (wheat), Sofor (smoky brown), and Waddeh (white). According to prior research, Mezayen breeds are often divided into two main groups: (1) the Majaheem breed alone and (2) the “Malaween” breeds, encompassing Homor, Shaele, Shageh, Sofor, and Waddeh breeds. Our main aim is to determine if we can find support for these two groups based on geometric morphometric analysis. We quantified the shape and size of each camel's torso based on landmarks and semilandmarks spanning their dorsolateral surface, from the root of the tail to the base of the neck. Landmarks and semilandmarks were digitized on photographs of 514 camels, sampled from > 23 breeders from various regions in Kuwait. The division of the Majaheem breed from the Malaween breeds was strongly supported. The Majaheem breed showed marked group homogeneity and significantly differed from all the Malaween breeds, while the Malaween breeds were largely heterogeneous, and did not significantly differ from each other. These results were supported by Procrustes ANOVA, between-group PCA, UPGMA cluster analysis, and Type-II ANOVA. Based on the TPS deformation grids, the Majaheem breed mainly differed from the Malaween breeds in the shape of the posterior part of the torso (the curvature from the tail root to the hump apex). We detected no significant allometry in shape variation, indicating no correlated variation of torso shape with size. This study demonstrates the usefulness of geometric morphometric analysis in camel breed classification. To utilize this method for camel breeding, further work is needed to determine whether our results could be generalized to other breeds, and to discover the combination of morphometric traits that best delineate breeds based on predefined breeding goals.