Tieo, S. | bioRxiv Science

bioRxiv2024

A deep learning approach to detect and visualise sexual dimorphism in monomorphic species

Sex recognition is facilitated by dimorphism in some traits. However, humans often fail to find the traits that allow to distinguish between sexes in other species. Deep learning has the potential to surpass humans in identifying cryptic differences between sexes, but, so far, has rarely been used to assess sexual dimorphism. In this study, we evaluated (i) the ability of a fine-tuned classification neural network, EfficientNet, to find differences between sexes in a species that appears monomorphic to humans, the sociable weaver (Philetairus socius). We then assessed (ii) the benefits of Grad-CAM visualisation techniques to understand which parts of the individuals are used by the network to differentiate the sexes. We trained 10-folds cross-validation models on more than 4,500 pictures of the head from more than 1,300 individuals. Our results show that the network can predict sex of sociable weavers with an accuracy of 76%, which is considerably higher than humans performance, and that the model was similarly good at predicting females and males. When interpreting the probability of being classified to one sex, our results further reveal an effect of the interaction of sex with age on the confidence score of the models which shows that younger males are less masculine than older ones, and older females more masculine than younger ones. Finally, using Grad-CAM we found that the model mostly uses the beak region to predict the sex of individuals. Overall, this work shows that artificial intelligence has the potential to be a non-invasive sexing tool, surpassing human capabilities and aiding in pinpointing potential cryptic dimorphic body parts that have yet to be identified. In birds, half of the worlds species appear sexually monomorphic to humans, and re-evaluation of species dimorphism with this type of methods could deepen our understanding of the effect of selection on animal traits.

zoology↗

bioRxiv2018

Quantitative analysis reveals the basic behavioural repertoire of the urochordate Ciona intestinalis.

Quantitative analysis of animal behaviour in model organisms is becoming an increasingly essential approach for tackling the great challenge of understanding how activity in the brain gives rise to behaviour. In addition, behavioural analysis can provide insight on the molecular basis of nervous system development and function as demonstrated by genetic screens focused on behavioural phenotyping in some genetically tractable model organisms. The progress in building low-cost automated tracking setups, together with advances in computer vision machine learning have expanded the repertoire of organisms which are amenable to quantitative behavioural analysis. Here we used automated image-based tracking to extract behavioural features from an organism of great importance in understanding the evolution of chordates, the free swimming larval form of the tunicate Ciona intestinalis which has a compact and fully mapped nervous system composed of only 231 neurons. We analysed hundreds of videos of larvae and we extracted basic geometric and physical descriptors of larval behaviour. Most importantly, we used machine learning methods to create an objective ontology of behaviours for C. intestinalis larvae. We identified eleven behavioural modes using agglomerative clustering. This approach enabled us to produce a quantitative description of the basic larval behavioural repertoire. Furthermore, we tested the robustness of this repertoire by comparing different rearing conditions and ages. Using our pipeline for quantitative behavioural analysis, we successfully reproduced the known photoresponsive behaviour and the first demonstration to our knowledge that C. intestinalis larvae exhibit sensory arousal and thigmotaxis, both of which can be modulated by the anxiotropic drug modafinil. Remarkably, by comparing the behaviour between animals assayed individually or in small groups, we found that crowd size influences larval behaviour. This study shows that C. intestinalis larval behaviour can be broken down to a set of stereotyped behaviours that are used to different extents in a context-dependent manner. Furthermore, it raises exciting possibilities such as mapping behaviour to specific neurons of this compact chordate nervous system and it paves the way for comparative quantitative behavioural studies as a means to reconstruct the evolution of behaviour, especially in the chordate lineage.

animal behavior and cognition↗