Srinivas C. Turaga

Chromatix provides a differentiable, GPU-accelerated wave-optics library to enable fast and parallelized optical simulations for computational optics applications.

The paper develops two variants of a glutamate indicator with enhanced sensitivity and tailored deactivation rates for studying synaptic transmission in vivo.

This paper develops a detailed biomechanical model of fruit flies for simulating locomotion using reinforcement learning.

This paper introduces a point affinity transformer framework for global neuron shape reasoning, improving segmentation error correction and neuron type classification in connectomics.

This paper introduces an auxiliary learning task using Local Shape Descriptors to improve neuron segmentation in electron microscopy volumes, achieving performance comparable to state-of-the-art methods with greater efficiency.

The paper creates a comprehensive atlas of the Drosophila larval nervous system using single-cell transcriptomics and validates findings with RNA-FISH, revealing gene expression changes linked to behavioral states.

The paper presents an automatic method to identify synaptic partners in a whole-brain Drosophila EM dataset using convolutional neural networks, achieving high accuracy in predicting synaptic connections.

The paper benchmarks spike rate inference algorithms for two-photon calcium imaging data and finds that top-performing algorithms show high correlation despite diverse approaches.

Deep learning rivals or beats baselines across imaging, health records, genomics and drug discovery but is not yet transformative; scarce labels, privacy and opacity remain hurdles.