NeuroElectro

The aim of the NeuroElectro Project is to represent structured electrophysiological information from diverse neuron types. Currently, this information is obtained by text-mining and manually curating the neuroscience literature (see the NeuroElectro Publications for details).

This repository encapsulates the majority of the features of the project, including text-mining (located in the folder labelled "article_text_mining"), the models specifying the relational database (implemented in Django), and the website front end and back end code (implemented in Python's Django web framework).

Interested collaborators and contributors should contact @stripathy or @rgerkin, post in the Gitter chat, or post to the Google Groups Mailing List.

Code Repository Objectives

• Information Management
  • Integrate and structure data related to neuron type electrophysiology, like resting membrane potentials or spike amplitudes
  • Develop and maintain APIs for accessing and downloading data
  • Keep data up-to-date
• Text-mining
  • Apply algorithms for downloading neuroscience articles
  • Identify electrophysiological parameters, neuron types and subtypes, and methodological information
  • Identify experimental factors like use of genetically modified animals
• Data Curation
  • Provide a curation interface for use by human curators to check and fix the text-mined content
• Web Interface
  • Provide an interactive web interface where extracted data is viewable and searchable
  • Ensure that all extracted data is trace-able back to its source

Some use cases

• I want to know the average value of an electrophysiological parameter in a neuron type
• I want to find publications reporting electrophysiological data about my favorite neuron type
• I need realistic parameters to constrain my computational model of a neuron type
• I want to see how variable an electrophysiological property (like resting membrane potential) is across neuron types and different publications within a neuron type.
• I want to see how experimental parameters and metadata affect electrophysiological measurements

• I want to compare electrophysiological variability across neuron types with gene expression variability