Eventer

Eventer detects spontaneous synaptic events acquired by electrophysiology or imaging using supervised learning to classify events and increase throughput and reproducibility in neuroscience research.

Key Features:

• Supervised learning classification: Random forests are created and trained to associate various features of events with manual labeling.
• Model training and application: The application learns the user's expert strategy for classifying detected event candidates from a small subset of data and automatically applies the same criterion to the remaining dataset.
• Model storage and reuse: The stored model file can be reloaded and used to analyse large datasets with greater consistency.
• Event candidate detection: Eventer first uses a suitable model template to pull out event candidates using fast Fourier transform (FFT)-based deconvolution with a low threshold.
• Model sharing repository: Researchers can upload and share their machine learning model files to enhance reproducibility when analyzing datasets of spontaneous synaptic activity.

Scientific Applications:

• Spontaneous synaptic event detection: Detection and analysis of spontaneous synaptic events from electrophysiology or imaging data in neuroscience research.
• Reproducible synaptic transmission analysis: Addressing reproducibility concerns in neuroscience research by replacing manual selection of candidate events with consistent, shareable machine learning criteria.
• High-throughput dataset analysis: Increasing throughput of data analysis for researchers studying synaptic transmission by applying trained models to large datasets.

Methodology:

Eventer uses fast Fourier transform (FFT)-based deconvolution with a low threshold and a suitable model template to pull out event candidates, then creates and trains random forests to associate various features of the events with manual labeling.