NEUROeSTIMator

NEUROeSTIMator quantifies neuronal activation from single-cell and spatial transcriptomic data using deep learning to estimate activity-dependent transcriptional responses underlying synaptic plasticity and neural circuit function.

Key Features:

• Integration of Transcriptomic Signals: Uses deep learning to integrate complex single-cell and spatial transcriptomic signals for estimating neuronal activation.
• Association with Electrophysiological Features: Activity estimates correlate with Patch-seq electrophysiological features.
• Robustness Across Variables: Produces estimates that are robust across species, cell types, and brain regions.
• Validation with Existing Studies: Validated against previously published single-cell activity-induced gene expression studies.
• Application in Spatial Transcriptomics: Applied to spatial transcriptomic data to reveal learning-induced neuronal activity patterns across brain regions in male mice.

Scientific Applications:

• Neuroscience Research: Investigating activity-dependent transcriptional processes related to synaptic plasticity, brain circuit development, behavioral adaptation, and long-term memory.
• Brain Circuit Analysis: Mapping activity patterns across brain regions to support studies of circuit development and function.
• Behavioral Studies: Linking neuronal activation estimates to learning-related and other behavioral adaptations.

Methodology:

Employs a deep learning framework that processes single-cell and spatial transcriptomic data at cellular resolution, integrates transcriptomic signals to estimate neuronal activity levels, and compares estimates to Patch-seq electrophysiological features and published activity-induced gene expression studies.