FOR INTERNAL USE ONLY – SAMPLE Q&A FOR MEDIA INTERVIEWS

BigNeuron Launch

BigNeuron Launch

FAQs

Updated 3.13.2015

Summary:

BigNeuron is a community effort to define and advance state-of-the-art of single neuron reconstruction: an essential unsolved challenge in brain science. The project will both standardize the methods to generate high quality and consistent data, and mobilize the reconstruction community to generate interest in solving these complex and interesting algorithmic problems.

Topline Messages:

  • BigNeuron will bench-test on a common open platform as many open-source, automated neuron reconstruction algorithms as possible, using very large scale, publicly available single 3D neuron image datasets acquired by several light microscopy methods.
  • The bench-test results will be produced using some of the fastest supercomputers in the world and then compared and validated using multiple criteria carefully defined by the computational neuroscience community.
  • The major results of BigNeuron will include:
  • a worldwide community-oriented and large single neuron morphology database,
  • a large set of open-source, community-based novel tools for neuroscience studies,
  • a standardized protocol and resource for any future scientists to study and reconstruct neuron morphology, and
  • a rich library of morphology feature definitions and algorithms to provide the basis of quality metrics and classification.
  • The Allen Institute for Brain Science is the main driver behind this international community effort.

Key Facts:

  • Scientific spokesperson: Hanchuan Peng, Ph.D.
  • Other spokespersons: Christof Koch, Ph.D., Jane Roskams, Ph.D.
  • Organizations involved in the collaboration (see below for more details):
  • Allen Institute for Brain Science (USA)
  • Janelia Research Campus (HHMI) (USA)
  • Oak Ridge National Laboratory (USA)
  • Lawrence Berkeley National Laboratory (USA)
  • Human Brain Project (Europe)
  • University of Cambridge (UK)
  • Wellcome Trust (UK)
  • NeuroMorpho.org and George Mason University (USA)
  • Web Intelligence Consortium and Brain Informatics conferences (Japan/China)
  • FlyCircuit.org (Taiwan)
  • Erasmus University Medical Center (Netherlands)

General Q&A

What is BigNeuron?

BigNeuron is a large-scale community project. The major goal of BigNeuron is to bench-test as many as possible open source, automated neuron reconstruction algorithms on a common and open platform for very large scale, single 3D neuron image datasets acquired using multi-modalities of light microscopy methods. The bench-test results will be produced using some of the fastest supercomputers in the world and then compared and validated using multiple criteria carefully defined by the computational neuroscience community.

Why do we need BigNeuron?

The three-dimensional shape of a neuron plays a major role in determining its connectivity, integration of synaptic input and cellular firing properties, and changes dynamically with its activity and state of the organism. Analyzing the 3D morphology of neurons in an unbiased way is critical to understand neuronal function, and developing applications to model circuitry.

Although advances in brain cellular imaging have yielded thousands of detailed images, there are no standards to collect and compare the data. BigNeuron aims to create those standards and determine which analysis approach is the most effective for the important questions driving the field.

Who is involved in BigNeuron, and what are their roles?

  • Allen Institute for Brain Science (USA)
  • Initialize and lead the project. Will coordinate the main effort, contribute neuron reconstruction algorithms and neuron data, sponsor and host the annotation workshop, lead data analysis and distribution effort.
  • Janelia Research Campus (HHMI) (USA)
  • Partner of the project. Will enable development of the Vaa3D platform and many tools, contribution of neuron reconstruction algorithms and neuron data, sponsor and host the USA/Canada hackathon.
  • Oak Ridge National Laboratory (USA)
  • Partner of the project. Will provide the major supercomputing facility.
  • Lawrence Berkeley National Laboratory (USA)
  • Partner of the project. Will provide some supercomputing facilities, potential data hosting (via CRCNS.org at Berkeley) and website support.
  • Human Brain Project (Europe)
  • Partner of the project. Co-sponsor of the Europe hackathon at Cambridge. Will provide some supercomputing facilities and contribute to data analysis.
  • University of Cambridge (UK)
  • The neuroscience department (PDN) and advanced imaging center will host the Cambridge hackathon.
  • Wellcome Trust (UK)
  • Partner of the project. Will co-sponsor the Europe hackathon at Cambridge.
  • NeuroMorpho.org and George Mason University (USA)
  • Partner of the project. Home institute of one of the main co-organizers (Giorgio Ascoli) for project planning, will contribute to data analysis, potentially host reconstruction data
  • Web Intelligence Consortium and Brain Informatics conferences (Japan/China)
  • Sponsor and host for the Asia/Australia hackathon, contribution of a neuron reconstruction algorithm.
  • FlyCircuit.org (Taiwan)
  • One of the main test data contributors, may also contribute one neuron reconstruction method
  • Erasmus University Medical Center (Netherlands)
  • Home institute of one of the co-organizers (Erik Meijering) who will participate in data analysis after bench testing and potentially also contribute one reconstruction algorithm.

What will be the outcome of BigNeuron?

The major anticipated results of the project are:

  • a world-wide community-oriented and large single neuron morphology database;
  • a large set of open-source, community-based novel tools for neuroscience studies;
  • a standardized protocol and resource for any future scientists who want to study neuron anatomy and neuron morphology reconstructions;
  • a rich library of morphology feature definitions and algorithms to provide the basis of quality metrics and classification.

How will BigNeuron help advance brain research worldwide?

BigNeuron will ultimately provide several key resources for scientists around the world.

  • A clear definition of which algorithms are most useful to analyze different types of neurons, saving time and allowing researchers even in disparate locations to effectively compare their results.
  • Standardized protocols so that researchers in different labs can reliably reproduce their results, making for more robust and meaningful data.
  • A rich library of images that scientists can draw on for their own investigations.

By creating clear standards, processes and metrics for digital reconstruction of single neurons, BigNeuron will foster collaboration and accelerate the pace of research into single neuron morphologies around the world.

What are the project phases of BigNeuron?

  • In Phase I (2015) of the project, a series of hackathons will be held in Beijing, Cambridge, and Janelia Research Campus (HHMI), cosponsored by a range of organizations. Neuron reconstruction algorithms contributed from participants will be ported onto a common software platform to analyze neuronal physical structure using the same core dataset. Additional 3D images will be solicited internationally at this time from multiple organisms.
  • In Phase II (2015-2016), all ported algorithms will be bench-tested at DOE (USA) and human brain project supercomputing centers, allowing us to standardize optimal protocols for labeling, visualizing and analyzing neuronal structure and key biological features. Computational annotation and analysis workshops will also be held at the Allen Institute (Seattle), Broad Institute (MIT) and in other ways, to help analyze the performance of neuron reconstruction.
  • In Phase III (2016-) the combined results of Phase I and II will be used to develop a comprehensive annotated database of complex neuronal morphology, generate a searchable tool for discovering annotated and unique characteristics of neuronal morphology, and lay the groundwork for potential integrating this with large-scale datasets linking form to neuronal function.

What is the relationship between BigNeuron and Neurodata Without Borders?

Both projects are aimed at creating standards for data collection and analysis, but that they address different issues. Neurodata Without Borders is a more general initiative, while BigNeuron is specifically about 3D neural reconstructions. BigNeuron also has a significant data-sharing component, since images from many different labs will be collected to create a database that researchers can access, along with the results of algorithm bench testing.

What kind of data will BigNeuron use? What types of algorithms?

The set of BigNeuron bench-test data will include neuron image stacks from different species (including fruit fly and other insects, fish, turtle, chicken, mouse, rat, and human) and nervous system regions such as cortical and subcortical areas, retina, and peripheral nervous system.The test data will be from multiple light microscopy modalities, especially laser scanning microscopy (confocal/2p) and brightfield or epi-fluorescent imaging.

Reconstruction visualization will be supported by Vaa3D, an open-source, cross-platform system for visualization and analysis. All algorithms to be bench tested must be ported into the Vaa3D platform.

What are the hackathons?

The planned hackathons will help contributors and developers port data and algorithms onto the common Vaa3D platform. Each hackathon will be one week long and largely focused on linking existing algorithms as executable Vaa3D plugins.

Is BigNeuron a crowd-sourcing project?

BigNeuron is a community project, but not a crowd-sourcing project. We are not asking people to annotate reconstructions. Instead, we ask developers and trained experts to provide different neuron reconstruction algorithms and available manual tracing and to join in for bench-testing and analysis in a collaborative and professional spirit. In the next phases of BigNeuron, there will be a crowd-sourcing and annotation component, to build the reference database as comprehensively as we can.

MARCH 2015alleninstitute.org

page1 of 4brain-map.org