Dorsoventral comparison of intraspecific variations in the butterfly wing pattern using a convolutional neural network

Butterfly wing patterns exhibit notable differences between the dorsal and ventral surfaces, and morphological analyses of them have provided insights into the ecological and behavioural characteristics of wing colour patterns. Conventional methods for dorsoventral comparisons are constrained by the need for homologous patches or shared features between two surfaces, limiting their applicability across species. We used a convolutional neural network (CNN)-based analysis, which can compare images of the two surfaces without focusing on homologous patches or features, to detect dorsoventral bias in intraspecific variations: sexual dimorphism and female-limited mimetic polymorphism (FMP). Using specimen images of 29 species, we first showed that the level of sexual dimorphism calculated by CNN-based analysis corresponded well with traditional assessments of sexual dissimilarity, demonstrating the validity of the method. Dorsal biases were widely detected in sexual dimorphism, suggesting th..., , , # Dorsoventral comparison of intraspecific polymorphisms in the butterfly wing pattern using a convolutional neural network

[https://doi.org/10.5061/dryad.djh9w0w8b](https://doi.org/10.5061/dryad.djh9w0w8b)

## Dataset (Butterfly specimen image dataset from \"Suguru Igarashi Insect Collection\")

#### File: butterflies\_dorsal\_ventral.zip

**Description:** Butterfly specimen images from 29 species observed in the Yaeyama Islands, Japan, were photographed from a large butterfly specimen dataset [\"Suguru Igarashi Insect Collection\"](https://umdb.um.u-tokyo.ac.jp/DEntomology/IgarashiInsect1/en/index.php), stored in the University Museum, The University of Tokyo (UMUT). Dorsal and ventral surfaces were photographed for each individuals. Please note that the images used in the analysis for the associated paper that are included in the preprint (DOI: [10.1101/2024.08.01.606114](https://www.biorxiv.org/content/10.1101/2024.08.01.606114v1)) have been removed, as they are not compatible with the...