Related Experiment Video
Updated: Aug 16, 2025
10:14
Tactile Conditioning And Movement Analysis Of Antennal Sampling Strategies In Honey Bees Apis mellifera L.
Published on: December 12, 2012
10.7K
Using the Software DeepWings© to Classify Honey Bees across Europe through Wing Geometric Morphometrics
Carlos Ariel Yadró García1,2, Pedro João Rodrigues2,3, Adam Tofilski4
1Centro de Investigação de Montanha, Instituto Politécnico de Bragança, Campus de Santa Apolónia, 5300-253 Bragança, Portugal.
Insects
|December 23, 2022
Summary
DeepWings© software accurately classifies honey bee subspecies using wing images. This machine learning tool shows strong performance on diverse datasets, aiding in subspecies identification and genetic integrity assessment.
Area of Science:
- Zoology
- Genetics
- Computer Science
Background:
- Accurate identification of honey bee subspecies is crucial for understanding population genetics and conservation efforts.
- Traditional methods for subspecies identification can be time-consuming and require specialized expertise.
- Geometric morphometrics offers a quantitative approach to analyzing biological shapes, including insect wings.
Purpose of the Study:
- To evaluate the performance of DeepWings©, a machine learning software for honey bee subspecies classification using wing geometric morphometrics.
- To test the software's accuracy on a large, diverse dataset of honey bee wing images from various sources.
- To assess the influence of genetic introgression on the accuracy of wing-based subspecies classification.
Main Methods:
- Utilized a dataset of 14,816 honey bee wing images representing five subspecies: A. m. carnica, Apis mellifera caucasia, A. m. iberiensis, Apis mellifera ligustica, and A. m. mellifera.
- Employed DeepWings© software, which uses machine learning algorithms to classify subspecies based on wing geometric morphometrics.
- Correlated wing classification results with molecular markers in a subset of colonies to assess genetic integrity.
Main Results:
- DeepWings© demonstrated good performance across various subspecies and image qualities, with high classification probabilities in native ranges (e.g., 92.6% for A. m. iberiensis, 88.3% for A. m. carnica).
- Lower accuracy was observed in introduced populations (Azores) and in areas with significant genetic introgression (A. m. mellifera range).
- A significant but weak association (r = 0.31) was found between wing morphometrics and molecular markers, influenced by C-derived introgression.
Conclusions:
- DeepWings© is a reliable tool for automated honey bee subspecies classification using wing images, even with variable data quality.
- The software's performance is affected by factors like geographic origin and genetic introgression, highlighting the complexity of honey bee population structures.
- Wing geometric morphometrics, when analyzed with machine learning, provides valuable insights into honey bee subspecies identification and population dynamics.
Related Concept Videos
01:28Methods of Classification and Identification
102
Bacterial identification relies on a diverse array of techniques to classify and understand microorganisms, each tailored to uncover specific characteristics. Traditional morphological approaches, while still valuable, are limited for closely related or structurally simple organisms. Modern methods integrate biochemical, serological, genetic, and advanced molecular tools to achieve greater accuracy.Morphological and Biochemical TechniquesMorphological characteristics, such as cell shape and...
102
02:54Evolutionary Relationships through Genome Comparisons
6.1K
Genome comparison is one of the excellent ways to interpret the evolutionary relationships between organisms. The basic principle of genome comparison is that if two species share a common feature, it is likely encoded by the DNA sequence conserved between both species. The advent of genome sequencing technologies in the late 20th century enabled scientists to understand the concept of conservation of domains between species and helped them to deduce evolutionary relationships across diverse...
6.1K