Biology subjects

Parkinson, R. H.

Publications and source records attributed to Parkinson, R. H..

6 recordsLinked to original sources

bioRxiv2026

BEEhaviourLab: A high-throughput platform for sublethal stressor screening in insects

Behavioural responses often provide the earliest detectable signs of physiological stress yet quantifying them at scale remains challenging. Acoustic behaviours such as buzzing are central to insect locomotion and communication but are rarely measured systematically, and their potential as sensitive indicators of environmental stress remains largely unexplored. Here, we test whether changes in insect acoustic behaviour can reveal sublethal effects of pesticide exposure. Using a newly developed automated behavioural platform (BEEhaviourLab), we recorded synchronised video and audio from insects, enabling simultaneous, high-throughput quantification of locomotor activity and buzzing behaviour across multiple individuals and species without the need for tagging. We applied this system to characterise the effects of the veterinary antiparasitic moxidectin on the bumble bee Bombus terrestris. Acute contact exposure produced dose-dependent reductions in locomotion and buzzing behaviour at concentrations below lethal thresholds. Acoustic measures detected behavioural disruption with sensitivity comparable to video-based activity metrics, demonstrating that buzzing behaviour provides a sensitive and previously unused indicator of neurotoxic stress. More broadly, scalable multimodal behavioural phenotyping enables subtle behavioural disruption to be detected across large experiments, opening new opportunities to incorporate acoustic endpoints into studies of animal behaviour, environmental stress, and ecological risk assessment.

animal behavior and cognition↗

bioRxiv2025

MetaBeeAI: an AI pipeline for full-text systematic reviews in biology

The volume and complexity of scientific literature are expanding rapidly, making it increasingly difficult to extract and synthesise information across studies. This challenge is particularly acute in the biological sciences, where evidence spans multiple levels of organisation and heterogeneous experimental designs. Large Language Model (LLM) pipelines offer a scalable route to evidence synthesis, but many existing approaches lack transparency, modularity, and effective mechanisms for human oversight. We present MetaBeeAI, an open-source, modular pipeline that integrates established LLM techniques into a coherent, auditable workflow for structured data extraction in biology. MetaBeeAI combines modular prompting, multi-pass extraction, and expert-in-the-loop validation within an interface that presents model outputs alongside source text, enabling inspection, correction, and iterative refinement. The pipeline produces machine-readable records of prompts, configurations, and expert annotations, supporting reproducibility and continuous improvement. We apply MetaBeeAI to 924 research papers on bees and pesticides, extracting structured information on species, compounds, exposure designs, and experimental context. Evaluation demonstrates improved consistency, convergence with expert judgement, and robustness across heterogeneous biological studies, highlighting the value of expert-guided refinement. MetaBeeAI provides a transparent and extensible framework for scalable evidence synthesis, supporting reliable integration of LLMs into biological research workflows.

ecology↗

bioRxiv2025

Bumblebees learn to use peripheral taste to predict the presence of nectar in flowers

Learning cues such as tastes associated with palatable food is an important mechanism animals have for foraging optimally. Insects can use gustatory receptor neurons (GRNs) in their mouthparts to detect nutrients and toxins, but they also taste compounds using sensilla on peripheral organs such as their antennae. Bees are adept at learning to associate floral traits with the presence of nectar rewards, but few studies have examined how they incorporate gustatory information from their antennae with rewards. Here, we characterize the ability of adult worker bumblebees (Bombus terrestris) to taste sugar, salt, and bitter compounds using their antennae and then tested whether they could use this sensory information to associate it with food. We show that bumblebees have antennal GRNs sensitive to sugars, salts, and bitter compounds and that they can use surface chemistry differences detected by their periphery to learn about the presence or absence of flower rewards in a free-flight assay. Naive bumblebees showed no instinctual preferences toward or against any surface chemistry tested. Bumblebees performed best when sucrose surface cues were associated with sucrose reward, but they could learn to associate any cue with the presence or absence of sucrose solution. Interestingly, the bees found it more difficult to associate quinine surface chemistry with the presence of reward than its absence. These results indicate that bees have the potential to learn to associate another floral trait - chemicals on the surfaces of petals - with the quality of floral rewards. Summary statementBehavioural experiments and electrophysiological recordings show bumblebees can detect peripheral taste cues on surfaces of artificial flowers, including bitter toxins, and learn to use these to predict rewards.

animal behavior and cognition↗

bioRxiv2024

Gustatory sensitivity to amino acids in bumblebees

Bees rely on amino acids from nectar and pollen for essential physiological functions. While nectar typically contains low (<1 mM) amino acid concentrations, while their levels in pollen are higher, but vary widely (10-200 mM). Behavioural studies suggest bumblebees have preferences for specific amino acids but whether such preferences are mediated via gustatory mechanisms remains unclear. This study explores bumblebees (Bombus terrestris) gustatory sensitivity to two essential amino acids (EAAs), valine and lysine, using electrophysiological recordings from gustatory sensilla on their mouthparts. Valine elicited a concentration-dependent response from 0.1 mM, indicating that bumblebees could perceive valine at concentrations found naturally in nectar and pollen. In contrast, lysine failed to evoke a response across tested concentrations (0.1-500 mM). The absence of lysine detection raises questions about the specificity and diversity of amino acid-sensitive receptors in bumblebees. Bees responded to valine at lower concentrations than sucrose, suggesting comparatively higher sensitivity (EC50: 0.7 mM vs. 3.91 mM for sucrose). Our findings indicate that bumblebees can rapidly evaluate the amino acid content of pollen and nectar using pre-ingestive cues, rather than relying on post-ingestive cues or feedback from their nestmates. Such sensory capabilities likely impact foraging strategies, with implications for plant-bee interactions and pollination.

neuroscience↗

bioRxiv2024

Do pollinators play a role in shaping the essential amino acids found in nectar?

O_LIPlants produce floral nectar as a reward for pollinators, which contains carbohydrates and amino acids (AAs). We designed experiments to test whether pollinators could exert selection pressure on the profiles of AAs in nectar. C_LIO_LIWe used HPLC to measure the free amino acids and sugars in the nectar of 102 UK plant species. Six distinct profiles of essential amino acids (EAAs) were defined using the relative proportions of AAs with a clustering algorithm; we then tested bumblebee (Bombus terrestris) preferences for the EAA profiles and proline using a two-choice assay. C_LIO_LIWe found a phylogenetic signal for the proportions of phenylalanine, methionine and proline as well as the total concentrations of essential and non-essential AAs. However, there was no phylogenetic signal for EAA profile. Bumblebees did not exhibit a preference for any of the six EAA nectar profiles, however, four of the EAA profiles stimulated feeding. In contrast, bumblebees avoided proline in an inverse concentration-dependent manner. C_LIO_LIOur data indicate that bees are likely to have mechanisms for the post-ingestive evaluation of free AAs in solution but are unlikely to taste EAAs at nectar-relevant quantities. We predict that EAAs increase nectar value to bumblebees post-ingestively. C_LI

ecology↗

bioRxiv2023

Bumblebee mouthparts exhibit poor acuity for the detection of pesticides in nectar

Bees are important pollinators of agricultural crops, but their populations are at risk when pesticides are used. One of the largest risks bees face is poisoning of floral nectar and pollen by insecticides. Studies of bee detection of neonicotinoids have reported contradictory evidence about whether bees can taste these pesticides in sucrose solutions and hence avoid them. Here, we use an assay for the detection of food aversion combined with single-sensillum electrophysiology to test whether the mouthparts of the buff-tailed bumblebee (Bombus terrestris) detect the presence of pesticides in a solution that mimicked the nectar of oilseed rape (Brassica napus). Bees did not avoid consuming solutions containing concentrations of imidacloprid, thiamethoxam, clothianidin, or sulfoxaflor spanning six orders of magnitude, even when these solutions contained lethal doses. Only extremely high concentrations of the pesticides altered spiking in gustatory neurons through a slight reduction in firing rate or change in the rate of adaptation. These data provide strong evidence that bumblebees cannot detect or avoid field-relevant concentrations of pesticides using information from their mouthparts. As bees rarely contact floral nectar with other body parts, we predict that they are at high risk of unwittingly consuming pesticides in the nectar of pesticide-treated crops.

neuroscience↗