Connectome-constrained networks predict neural activity across the fly visual system

Nature, Год журнала: 2024, Номер 634(8036), С. 1132 - 1140

Опубликована: Сен. 11, 2024

Язык: Английский

---

A Drosophila computational brain model reveals sensorimotor processing

Nature, Год журнала: 2024, Номер 634(8032), С. 210 - 219

Опубликована: Окт. 2, 2024

Язык: Английский

---

The fly connectome reveals a path to the effectome

Nature, Год журнала: 2024, Номер 634(8032), С. 201 - 209

Опубликована: Окт. 2, 2024

Язык: Английский

---

Biophysical neural adaptation mechanisms enable artificial neural networks to capture dynamic retinal computation

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Июль 16, 2024

Abstract Adaptation is a universal aspect of neural systems that changes circuit computations to match prevailing inputs. These facilitate efficient encoding sensory inputs while avoiding saturation. Conventional artificial networks (ANNs) have limited adaptive capabilities, hindering their ability reliably predict output under dynamic input conditions. Can embedding mechanisms in ANNs improve performance? To answer this question, we develop new deep learning model the retina incorporates biophysics photoreceptor adaptation at front-end conventional convolutional (CNNs). CNNs build on ’Deep Retina,’ previously developed retinal ganglion cell (RGC) activity. include layer outperform CNN models predicting male and female primate rat RGC responses naturalistic stimuli local intensity large ambient illumination. improved predictions result directly from within phototransduction cascade. This research underscores potential using them determine how circuits manage complexities natural are span range light levels.

Язык: Английский

---

Artificial intelligence for life sciences: A comprehensive guide and future trends

The Innovation Life, Год журнала: 2024, Номер unknown, С. 100105 - 100105

Опубликована: Янв. 1, 2024

<p>Artificial intelligence has had a profound impact on life sciences. This review discusses the application, challenges, and future development directions of artificial in various branches sciences, including zoology, plant science, microbiology, biochemistry, molecular biology, cell developmental genetics, neuroscience, psychology, pharmacology, clinical medicine, biomaterials, ecology, environmental science. It elaborates important roles aspects such as behavior monitoring, population dynamic prediction, microorganism identification, disease detection. At same time, it points out challenges faced by application data quality, black-box problems, ethical concerns. The are prospected from technological innovation interdisciplinary cooperation. integration Bio-Technologies (BT) Information-Technologies (IT) will transform biomedical research into AI for Science paradigm.</p>

Язык: Английский

---

NeuroMechFly 2.0, a framework for simulating embodied sensorimotor control in adultDrosophila

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2023, Номер unknown

Опубликована: Сен. 18, 2023

Abstract Discovering principles underlying the control of animal behavior requires a tight dialogue between experiments and neuromechanical models. Until now, such models, including NeuroMechFly for adult fly, Drosophila melanogaster , have primarily been used to investigate motor control. Far less studied with realistic body models is how brain systems work together perform hierarchical sensorimotor Here we present v2, framework that expands modeling by enabling visual olfactory sensing, ascending feedback, complex terrains can be navigated using leg adhesion. We illustrate its capabilities first constructing biologically inspired locomotor controllers use feedback path integration head stabilization. Then, add sensing this controller train it reinforcement learning multimodal navigation task in closed loop. Finally, more biorealistic two ways: our model navigates odor plume taxis strategy, uses connectome-constrained system network follow another simulated fly. With framework, accelerate discovery explanatory nervous develop machine learning-based autonomous artificial agents robots.

Язык: Английский

---

Building egocentric models of local space from retinal input

Current Biology, Год журнала: 2024, Номер 34(23), С. R1185 - R1202

Опубликована: Дек. 1, 2024

Язык: Английский

---

Morphology and synapse topography optimize linear encoding of synapse numbers inDrosophilalooming responsive descending neurons

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Апрель 28, 2024

ABSTRACT Synapses are often precisely organized on dendritic arbors, yet the role of synaptic topography in integration remains poorly understood. Utilizing electron microscopy (EM) connectomics we investigate Drosophila melanogaster looming circuits, focusing retinotopically tuned visual projection neurons (VPNs) that synapse onto descending (DNs). a given VPN type project to non-overlapping regions DN dendrites. Within these spatially constrained clusters, synapses not organized, but instead adopt near random distributions. To how this organization strategy impacts integration, developed multicompartment models DNs fitted experimental data and using precise EM morphologies locations. We find dendrite normalize EPSP amplitudes individual inputs distributions ensure linear encoding numbers from VPNs. These findings illuminate influences suggest may be default established through connectivity passive neuron properties, upon which active properties plasticity can then tune as needed.

Язык: Английский

---

AI networks reveal how flies find a mate

Nature, Год журнала: 2024, Номер 629(8014), С. 1010 - 1011

Опубликована: Май 22, 2024

Язык: Английский

---

Predicting natural behaviour by perturbation

Nature reviews. Neuroscience, Год журнала: 2024, Номер 25(8), С. 516 - 516

Опубликована: Июль 2, 2024

Язык: Английский

---