Cluster-Based Assessment of Growth Performance in Cameroon Local Chicken

Abstract

This study integrates clustering and network analyses to identify performance-based groups and their connections to phenotypes, providing insights to zoptimize local chicken breeding programs in Cameroon. A total of 113 Birds were grouped based on growth patterns between 16 and 22 weeks, assessed using total weight gain (TWG) and leg circumference gain (LCG). Repeated measures and Welch ANOVA were used to test performance differences, while Generalized Least Squares (GLS) ANCOVA identified growth predictors. Although phenotypic diversity was observed, TWG and LCG did not differ significantly across phenotypes (p > 0.05), whereas final body weight (BW) and leg circumference (LC) at 22 weeks did (p<0.01). Cluster analysis identified four distinct performance cluster groups independent of phenotype, with significant divergence in performance. Birds in Clusters 3 and 4 consistently outperformed (p<0.001) their counterparts in the other Clusters 1 and 2, showing superior TWG and LCG while sexual dimorphism was in favour of males (p < 0.05). Cluster 4 exhibited the highest final BW, whereas Cluster 3 had the greatest total weight gain, indicating a distinct tendency for early and late-stage growth, which could be strategically zoptimized for selective crossbreeding to combine their complementary traits. Network analysis indicates historical gene flow and possible heterozygosity within the population, with Normal and Feathered shank phenotypes potentially serving as genetic bridges for performance traits, while the distinct peripheral positioning of Feathered leg and Naked neck, linked only to Clusters 1 and 2, suggests genetic distinctiveness. GLS-ANCOVA confirmed Cluster 3 and 4, as the most significant predictors of TWG (p<0.001), alongside sex (p<0.05) and LCG (p<0.05). Integrating cluster and network analysis can enhance sustainable breeding strategies in low-input systems, balancing growth efficiency with genetic diversity. Breeders and policymakers are encouraged to adopt systematic performance recording practices and promote cross-cluster crossbreeding within local flocks.