Landmark-based truss morphometrics and COI barcoding reveal differentiation among Channa species and Channa striata populations in Bangladesh.

Accurate identification of Channa species and distinction between native and cultured stocks are important for both aquaculture and conservation in Bangladesh. This study combined landmark-based truss morphometrics with mitochondrial cytochrome c oxidase subunit I (COI) barcoding to examine morphological and mitochondrial differentiation among selected Channa species, with particular attention to native Bangladeshi and Vietnam-originated cultured C. striata. Morphometric variation was assessed using 13 homologous landmarks and 26 truss measurements after allometric size standardization, followed by principal component analysis, multivariate analysis of variance, and UPGMA clustering. COI sequences from 21 specimens were analysed together with reference sequences using genetic-distance, substitution-pattern, maximum-likelihood, and maximum-parsimony approaches. The morphometric analyses revealed clear body-shape differences among the examined taxa and between the two C. striata populations. The COI data also separated the major species groups and consistently distinguished native Bangladeshi from Vietnam-originated cultured C. striata. The representative genetic distance between the two C. striata populations was 0.0341, while within-group distances were 0.0441 for the native population and 0.0005 for the cultured stock. Maximum-likelihood and maximum-parsimony trees were generally consistent at the species and population levels, although deeper relationships were less stable and received weaker support. Morphological and molecular clustering were therefore not fully congruent, likely because body shape can be influenced by environmental conditions and phenotypic plasticity, whereas COI reflects maternal lineage history. Overall, the findings support clear morphological and mitochondrial differentiation among the examined Channa taxa and between the two C. striata populations. Broader geographic sampling and the inclusion of nuclear or genome-wide markers will be needed to assess admixture, local adaptation, and population structure more fully.