CEBRA

About CEBRA

CEBRA is a groundbreaking platform designed for neuroscience research, enabling the discovery of latent embeddings from simultaneous behavioral and neural data. Its innovative method accurately models neural dynamics during adaptive behaviors, revealing hidden structures in data and aiding researchers in hypothesis testing and complex behavior analysis.

CEBRA offers open access to its algorithm, with no subscription costs for users. Researchers can freely utilize its code available on GitHub, maximizing its benefits without any financial barriers. Such accessibility promotes collaborative scientific advancement and enhances effective use of behavioral and neural datasets.

CEBRA features an intuitive interface designed for seamless user experience, streamlining the process of dataset integration and analysis. Its layout allows users to efficiently access tools for neural data decoding and behavioral analysis, ensuring high usability and enhanced productivity in research applications.

How CEBRA works

Users start by onboarding CEBRA with their behavioral and neural datasets. After integrating data, they navigate its user-friendly interface to select analysis options—either hypothesis-driven or discovery-driven. The platform leverages powerful machine learning techniques to produce latent embeddings and decode neural activity, ultimately revealing deep insights into adaptive behaviors.

Key Features for CEBRA

Joint Behavioral and Neural Data Analysis

CEBRA excels in analyzing joint behavioral and neural data, offering researchers a powerful tool to uncover hidden patterns. By producing high-performance latent embeddings, it enables scientists to decode neural activity during complex behaviors, enhancing our understanding of neural dynamics and behavior correlations.

High-Performance Decoding

CEBRA's high-performance decoding feature allows researchers to interpret neural activity with exceptional accuracy. This capability is especially beneficial in visual cortex studies, enabling the reconstruction of viewed videos based on decoded neural signals, thus providing invaluable insights into sensory processing in real-time.

Multi-Session and Label-Free Analysis

CEBRA supports multi-session datasets and offers a label-free analysis option, catering to diverse research needs. This flexibility allows researchers to leverage extensive data over time or engage in exploratory analysis, making it a versatile tool for advancing neuroscience research and behavioral studies.