Large deformation diffeomorphic mapping of 3D shape variation reveals two distinct mandible and head capsule morphs in Atta vollenweideri leaf-cutter worker ants

Ants are crucial ecosystem engineers, and their ecological success is facilitated by a division of labour among sterile "workers". In some ant lineages, workers have undergone further morphological differentiation, resulting in differences in body size, shape, or both. Distinguishing between changes in size and shape is not trivial. Traditional approaches based on allometry reduce complex 3D shapes into simple linear, areal, or volume metrics; modern approaches using geometric morphometrics typically rely on landmarks, introducing observer bias and a trade-off between effort and accuracy. Here, we use a landmark-free method based on large deformation diffeomorphic metric mapping (LDDMM) to assess the co-variation of size and 3D shape in the mandibles and head capsules of Atta vollenweideri leaf-cutter ants, a species exhibiting extreme worker size-variation. Body mass varied by more than two orders of magnitude, but a shape atlas created via LDDMM on μ-CT-derived 3D mesh files revealed only two distinct head capsule and mandibles shapes-one for the minims (body mass < 1 mg) and one for all other workers. We discuss the functional significance of the identified 3D shape variation, and its implications for the evolution of extreme polymorphism in Atta.

replicAnt: a pipeline for generating annotated images of animals in complex environments using Unreal Engine

Deep learning-based computer vision methods are transforming animal behavioural research. Transfer learning has enabled work in non-model species, but still requires hand-annotation of example footage, and is only performant in well-defined conditions. To help overcome these limitations, we developed replicAnt, a configurable pipeline implemented in Unreal Engine 5 and Python, designed to generate large and variable training datasets on consumer-grade hardware. replicAnt places 3D animal models into complex, procedurally generated environments, from which automatically annotated images can be exported. We demonstrate that synthetic data generated with replicAnt can significantly reduce the hand-annotation required to achieve benchmark performance in common applications such as animal detection, tracking, pose-estimation, and semantic segmentation. We also show that it increases the subject-specificity and domain-invariance of the trained networks, thereby conferring robustness. In some applications, replicAnt may even remove the need for hand-annotation altogether. It thus represents a significant step towards porting deep learning-based computer vision tools to the field.

Learning from algorithm-generated pseudo-annotations for detecting ants in videos

Deep learning (DL) based detection models are powerful tools for large-scale analysis of dynamic biological behaviors in video data. Supervised training of a DL detection model often requires a large amount of manually-labeled training data which are time-consuming and labor-intensive to acquire. In this paper, we propose LFAGPA (Learn From Algorithm-Generated Pseudo-Annotations) that utilizes (noisy) annotations which are automatically generated by algorithms to train DL models for ant detection in videos. Our method consists of two main steps: (1) generate foreground objects using a (set of) state-of-the-art foreground extraction algorithm(s); (2) treat the results from step (1) as pseudo-annotations and use them to train deep neural networks for ant detection. We tackle several challenges on how to make use of automatically generated noisy annotations, how to learn from multiple annotation resources, and how to combine algorithm-generated annotations with human-labeled annotations (when available) for this learning framework. In experiments, we evaluate our method using 82 videos (totally 20,348 image frames) captured under natural conditions in a tropical rain-forest for dynamic ant behavior study. Without any manual annotation cost but only algorithm-generated annotations, our method can achieve a decent detection performance (77% in [Formula: see text] score). Moreover, when using only 10% manual annotations, our method can train a DL model to perform as well as using the full human annotations (81% in [Formula: see text] score).

Developmental biomechanics and age polyethism in leaf-cutter ants

Many social insects display age polyethism: young workers stay inside the nest, and only older workers forage. This behavioural transition is accompanied by genetic and physiological changes, but the mechanistic origin of it remains unclear. To investigate if the mechanical demands on the musculoskeletal system effectively prevent young workers from foraging, we studied the biomechanical development of the bite apparatus in Atta vollenweideri leaf-cutter ants. Fully matured foragers generated peak in vivo bite forces of around 100 mN, more than one order of magnitude in excess of those measured for freshly eclosed callows of the same size. This change in bite force was accompanied by a sixfold increase in the volume of the mandible closer muscle, and by a substantial increase of the flexural rigidity of the head capsule, driven by a significant increase in both average thickness and indentation modulus of the head capsule cuticle. Consequently, callows lack the muscle force capacity required for leaf-cutting, and their head capsule is so compliant that large muscle forces would be likely to cause damaging deformations. On the basis of these results, we speculate that continued biomechanical development post eclosion may be a key factor underlying age polyethism, wherever foraging is associated with substantial mechanical demands.