Outline-based geometric morphometrics: Wing cell differences for mosquito vector classification in the Tanaosri mountain range, Thailand.

Recent studies have revealed taxonomic signals within the wing cells of certain mosquito species. In our study, wing cell differentiation among mosquito vectors from the Tanaosri mountain range in Thailand was evaluated using the outline-based geometric morphometric (GM) approach. Our focus was on four specific wing cells for GM analysis: the wing contour (external cell), the second submarginal cell (internal cell 1), the first posterior cell (internal cell 2), and the third posterior cell (internal cell 3). Before proceeding with the GM approach, the identity of seven mosquito genera and 21 species was confirmed using molecular techniques. Our validated classification tests demonstrated that the performance of mosquito species classification varies according to genus. Notably, three Aedes species exhibited the highest accuracy for both internal cell 2 and internal cell 3, each registering a score of 93.20 %. In the case of two Mansonia species, the wing contour displayed a remarkable accuracy of 98.57 %. Consequently, we suggest the use of the outline-based GM approach, particularly focusing on the wing contour, for differentiating Mansonia annulifera and Mansonia uniformis. In contrast, the highest accuracy for classifying Culex species was found in internal cell 1, at 75.51 %, highlighting the challenges due to similarities in wing cells within this genus. These findings provide a guideline for future applications of the outline-based GM approach, focusing on wing cells, as an alternative method to classify mosquito vector species.