Biology subjects

Rapp, H.

Publications and source records attributed to Rapp, H..

5 recordsLinked to original sources

bioRxiv2025

Larvaworld : A behavioral simulation and analysis platform for Drosophila larva

Behavioral modeling supports theory building and evaluation across disciplines. Leveraging advances in motion-tracking and computational tools, we present a virtual laboratory for Drosophila larvae that integrates agent-based modeling with multiscale neural control and supports analysis of both simulated and experimental data. Virtual larvae are implemented as 2D agents capable of realistic locomotion, guided by multimodal sensory input and constrained by a dynamic energy-budget model that balances exploration and exploitation. Each agent is organized as a hierarchical, behavior-based control system comprising three layers: low-level locomotion, optionally incorporating neuromechanical models; mid-level sensory processing; and high-level behavioral adaptation. Neural control models can range from simple linear transfer models to rate-based or spiking neural network models, e.g. to accomodate associative learning. Simulations operate across sub-millisecond neuronal dynamics, sub-second closed-loop behavior, and circadian-scale metabolic regulation. Users can configure both larval models and virtual environments, including sensory landscapes, nutrient sources, and physical arenas. Real-time visualization is integrated into the simulation and analysis pipeline, which also allows for standardized processing of motion-tracking data from real experiments. Distributed as an open-source Python package, the platform includes tutorial experiments to support accessibility, customization, and use in both research and education. Author summaryLarvaworld was developed to address two key challenges in behavioral neuroscience and computational modeling. First, it responds to the growing call for closer collaboration between experimentalists and modelers by providing a shared platform -a virtual laboratory-where experimental data analysis and behavioral modeling can be seamlessly integrated. By standardizing dataset formats and ensuring identical, unbiased analysis pipelines for experimental and simulated data, Larvaworld facilitates methodological consistency and enables rigorous model evaluation. Second, it aims to bridge a long-standing gap in theory building and computational modeling at the level of the individual behaving organism. Historically, neuroscience has focused on sub-individual processes, while ecology has concentrated on supra-individual dynamics, resulting in discontinuities among the respective modeling approaches. Recent advances, however, have begun to align these fields, with neuroscience incorporating slower homeostatic processes and ecology integrating faster neurally-mediated mechanisms. Larvaworld boosts this convergence by adopting a nested, multi-timescale modeling approach, thus achieving behavioral regulation within the normative homeostatic constraints as these dynamically unfold during larval development. By combining established modeling paradigms from neuroscience and ecology, it provides a novel and flexible platform for studying behavior at the level of the individual organism, promoting cross-disciplinary insights and advancing computational neuroethology [1].

neuroscience↗

bioRxiv2024

In silico PK predictions in Drug Discovery: Benchmarking of Strategies to Integrate Machine Learning with Empiric and Mechanistic PK modelling

A successful drug needs to combine several properties including high potency and good pharmacokinetic (PK) properties to sustain efficacious plasma concentration over time. To estimate required doses for preclinical animal efficacy models or for the clinics, in vivo PK studies need to be conducted. While the prediction of ADME properties of compounds using Machine Learning (ML) models based on chemical structures is well established in drug discovery, the prediction of complete plasma concentration-time profiles has only recently gained attention. In this study, we systematically compare various approaches that integrate ML models with mechanistic PK models to predict PK profiles in rats after i.v. administration prior to synthesis. More specifically, we compare a standard noncompartmental analysis (NCA) based approach (prediction of CL and Vss), a pure ML approach (non-mechanistic PK description), a compartmental modeling approach, and a physiologically based pharmacokinetic (PBPK) approach. Our study based on internal preclinical data shows that the latter three approaches yield PK profile predictions of comparable accuracy (evaluated as geometric mean fold errors for each profile) across a large test set (>1000 small molecules). In summary, we demonstrate the improved ability to prioritize drug candidates with desirable PK properties prior to synthesis with ML predictions.

bioinformatics↗

bioRxiv2020

A spiking neural program for sensory-motor control during foraging in flying insects.

Foraging is a vital behavioral task for living organisms. Behavioral strategies and abstract mathematical models thereof have been described in detail for various species. To explore the link between underlying neural circuits and computational principles we present how a biologically detailed neural circuit model of the insect mushroom body implements sensory processing, learning and motor control. We focus on cast & surge strategies employed by flying insects when foraging within turbulent odor plumes. Using a spike-based plasticity rule the model rapidly learns to associate individual olfactory sensory cues paired with food in a classical conditioning paradigm. We show that, without retraining, the system dynamically recalls memories to detect relevant cues in complex sensory scenes. Accumulation of this sensory evidence on short time scales generates cast & surge motor commands. Our generic systems approach predicts that population sparseness facilitates learning, while temporal sparseness is required for dynamic memory recall and precise behavioral control. Our work successfully combines biological computational principles with spike-based machine learning. It shows how knowledge transfer from static to arbitrary complex dynamic conditions can be achieved by foraging insects and may serve as inspiration for agent-based machine learning.

neuroscience↗

bioRxiv2020

Neural computation underlying rapid learning and dynamic memory recall for sensori-motor control in insects.

Foraging is a vital behavioral task for living organisms. Behavioral strategies and abstract mathematical models thereof have been described in detail for various species. To explore the link between underlying nervous systems and abstract computational principles we present how a biologically detailed neural circuit model of the insect mushroom body implements sensory processing, learning and motor control. We focus on cast & surge strategies employed by flying insects when foraging within turbulent odor plumes. Using a synaptic plasticity rule the model rapidly learns to associate individual olfactory sensory cues paired with food in a classical conditioning paradigm. Without retraining, the system dynamically recalls memories to detect relevant cues in complex sensory scenes. Accumulation of this sensory evidence on short timescales generates cast & surge motor commands. Our systems approach is generic and predicts that population sparseness facilitates learning, while temporal sparseness is required for dynamic memory recall and precise behavioral control.

neuroscience↗

bioRxiv2019

Estimating numerosity of visual concepts with a single spiking neuron.

Insects are able to solve basic numerical cognition tasks. We show that estimation of numerosity can be realized and learned by a single spiking neuron with an appropriate synaptic plasticity rule. This model can be efficiently trained to detect arbitrary spatio-temporal spike patterns on a noisy and dynamic background with high precision and low variance. When put to test in a task that requires counting of visual concepts in a static image it required considerably less training epochs than a convolutional neural network to achieve equal performance. When mimicking a behavioral task in free flying bees that requires numerical cognition the model reaches a similar success rate in making correct decisions. We propose that using action potentials to represent basic numerical concepts with a single spiking neuron is beneficial for organisms with small brains and limited neuronal resources.

neuroscience↗