Sharmishtaa Seshamani

3D-MAESTRO automates processing of large-scale volumetric brain microscopy data, enabling scalable and reproducible analysis of mouse brain structures.

The paper identifies topographic organization in locus coeruleus norepinephrine neurons, linking their morphology, gene expression, and activity to specific brain regions and learning-related behaviors.

The paper investigates the synaptic connectivity of thick tufted layer 5 pyramidal neurons in the mouse visual cortex, revealing their local inhibitory and long-range excitatory connections.

The paper predicts MET-type identities from EM data and shows they have distinct connectivity patterns and synapse targets.

Electron microscopy of 1352 neurons in a mouse visual cortex column reveals widespread cell-type-specific inhibition, a basket-cell-targeting disinhibitory class, and interneuron motif groups sharing excitatory targets.

The paper presents a scalable pipeline for assembling large electron microscopy datasets into 3D volumes, achieving high throughput and modular design for efficient processing.

The paper uses electron microscopy data to classify cell types in the neocortex based on perisomatic features, showing that these features are sufficient for accurate classification even in large datasets.

The paper identifies oligodendrocyte precursor cells (OPCs) as major contributors to axon pruning in the mouse neocortex through phagocytic activity.

The paper analyzes single-cell RNA-seq data from mouse neocortical neurons to identify widespread neuropeptide signaling networks across all cortical neurons.

The paper introduces a method to predict fluorescently labeled subcellular structures from 3D brightfield images without using fluorescent dyes, demonstrating its use in static and time-lapse imaging.