Biology subjects

Stöckl, A.

Publications and source records attributed to Stöckl, A..

3 recordsLinked to original sources

bioRxiv2026

sstar2: A Python Package for S*-based Archaic Introgression Detection with Machine Learning

Detecting introgressed genomic fragments from unsampled or extinct source populations remains challenging. The S* statistic is widely used for this purpose, but the original sstar implementation relies on generalized additive models to smooth quantile-specific values precomputed from fixed count bins, requiring simulations with fixed numbers of segregating sites. Here, we present sstar2, a Python update that replaces this procedure with quantile regression to directly estimate S* thresholds at specified null quantiles from simulated genomic windows. We benchmarked sstar2 against the original sstar, linear quantile regression, and random forest quantile regression across three demographic models with both phased and unphased simulated data. sstar2 showed the best overall performance among the evaluated methods, with the most pronounced improvement under a challenging demographic model of ghost introgression in bonobos. These results show that sstar2 improves S* threshold calibration while making S*-based introgression analyses more flexible and compatible with modern simulation workflows.

bioinformatics↗

bioRxiv2026

Conservation of a lateralized visuo-motor axis in hawkmoth proboscis probing

Lateralization of behaviour, including motor control and sensory processing, is widespread across bilaterians. In visually guided tasks it often manifests as an axis aligning eye, appendage, and a target within a shared reference frame, such as eye-hand coordination in humans or eye-beak coordination in birds. While studied intensively in a few vertebrate systems, whether similar control principles apply to invertebrates, and more generally, how sensory and motor lateralization are linked mechanistically, remains unclear. Using the proboscis inspection behaviour of hummingbird hawkmoth Macroglossum stellatarum as a model for visual appendage guidance, our study provides evidence for lateralized visuo-motor control in an invertebrate. Combining high-speed videography and markerless pose estimation, we establish the underlying control axis between the hawkmoths unpaired appendage and its eye. We demonstrate that individuals displayed stable, idiosyncratic proboscis lateralization, which was tightly linked to their instantaneous viewing angle of visual targets, thus forming a persistent eye- proboscis-target axis. This axis also had functional consequences for feature-targeting. Assessing the sensory-motor plasticity using monocular occlusion, we found that moths preserved their lateralized visuo-motor geometry by adjusting body posture during flower inspection. Our findings suggest convergent control principles with vertebrate models of lateralized visual appendage guidance, while highlighting stark differences in sensory-motor plasticity, thus adding to our general understanding of how lateralization shapes control strategies across nervous systems. Significance StatementLateralization is widespread across animals and shapes how sensation and action are coordinated. Visually guided reaching with appendages is frequently lateralized across taxa, reflecting biases like handedness and eye dominance. However, the mechanistic link between lateralized sensing and motor control, and the extent of their plasticity, remain poorly understood, particularly in invertebrates. Here we show that the hummingbird hawkmoth integrates individually lateralized vision and proboscis probing movements into a unified control axis. Upon sensory perturbation, moths adjusted their position and body posture to maintain this axis. The visuo-motor lateralization produced measurable functional consequences, revealing it as a form of sensory-motor optimization. These findings uncover convergent principles of lateralized visuo-motor control across insects, humans, birds, and elephants.

animal behavior and cognition↗

bioRxiv2025

Hawkmoths learn best without negative reinforcement in a differential colour association task

Visual learning in insects can be strongly influenced by the underlying context and conditions. Improved colour learning has been demonstrated with differential conditioning using reward-aversion paradigms in eusocial insects like wasps, bumblebees, and honeybees, where substances that induce strong aversion, such as quinine, enable more accurate learning. Although the associations created during learning are intrinsically linked to the reinforcements used, few studies have compared the role of negative reinforcements in shaping these associations, and their relevance across insect groups with different life histories. In our study, we compared the effects of aversive substances for learning in the hummingbird hawkmoth (Macroglossum stellatarum), a solitary pollinator that relies on vision for foraging. Linking to previous learning studies in insects, we combined a sugar-rewarded target with a distractor that was paired with either quinine, salt, citric acid, water, or was presented with no aversive substance. Learning was assessed by conditioning hawkmoths to invert a strong colour preference between perceptually close or distant colour pairs, to provide tasks with disparate challenges. Contrary to results from eusocial insect species, hawkmoths trained with quinine were worse at switching preferences between similar colours compared to training with citric acid or appetitive-only differential conditioning. Learning success was associated with the animals foraging plasticity, where negative reinforcements were found to suppress exploration during foraging. Furthermore, we show that quinine interfered with sucrose perception, potentially impairing target acquisition during conditioning. Our results provide insights into the impact of negative reinforcements on solitary foragers and highlight the role of sensory ecology and life history in shaping learning outcomes.

animal behavior and cognition↗