D3171: Classification of repeated behavioural motifs of Aedes aegypti

Aedes aegypti have adapted incredibly to spread across all continents. This was partly possible since Ae. aegypti thrive in urban areas where human activity is at its peak. The success of this species has been attributed to its biological and behavioural adaptation to utilize urban environment to produce and lay eggs. Therefore, understanding this species baseline behaviour can give insight to the behavioural elements that drive their huge success at their spread. To this end, we attempt to statistically classify and model the adult stage of the mosquito species. All the behavioural experiments will be captured by Raspberry Pi cameras. To ensure precise tracking of the mosquito movements, DeepLabCut will be employed. DeepLabCut is a convolutional neural network that does markerless pose-estimation permitting for capturing all aspects of an animal’s motor movements. This intensive quantitative behavioural study permits us to have further grasp of Ae. aegypti behaviour which, in turn, can be used to implement effective vector control methods.