CONP Portal | About

This portal is a web interface for the Canadian Open Neuroscience Platform (CONP) to facilitate open science in the neuroscience community. CONP simplifies global researcher access and sharing of datasets and tools. The portal internalizes the cycle of a typical research project: starting with data acquisition, followed by processing using already existing/published tools, and ultimately publication of the obtained results including a link to the original dataset.

This infrastructure was built using technologies and best practices employed to make sharing easier and reproducible. Datasets are tracked and indexed using DataLad and Git-Annex. To ensure reproducibility of results, tools and pipelines can be run using Boutiques and a container engine ( Docker or Singularity ). In addition, some pipelines can also be run using High Performance Computing (HPC) via hooks to the CBRAIN platform.

More information about CONP can be found here. A tutorial on how to use the CONP portal can be found here.

What are the portal capabilities?

The portal is a conduit that incorporates best practices for data sharing and reproducibility. The graph below highlights the portal capabilities.

CONP graph

In the future, the portal will offer streamlined data download capabilities and an interface to facilitate the creation of the dataset descriptor ( DATS.json ). Additional analytics and advanced querying (e.g. Nexus ) will also be available. Finally, we are working on integrating citation capabilities and data publishing of Notebooks via Neurolibre.

What kinds of data can I find on the CONP Portal?

The CONP Portal provides a gateway to a diverse range of 40 datasets, primarily neuroimaging but also transcriptomics, genomics, and other related data modalities. Many of these datasets have been made accessible by neuroscience research institutes across Canada, while others link to publicly available resources that may be of interest to neuroscientists. A full list can be found here.

Dataset Graph

One of the key dataset available through the portal is the PREVENT-AD open dataset that includes longitudinal MRI follow-up and some basic demographics on a population of individuals with a family history of Alzheimer’s Disease. In the near future, additional data will be shared via a registered access for this PREVENT-AD cohort with clinical, cognitive and other biomarkers measures that were collected on this rich cohort. The original PREVENT-AD dataset is available here. The same dataset organized according to the BIDS standard can be found here.

How can I access datasets

CONP allows you to access and process data on public and private infrastructures via the CONP Portal. Data stored in private projects can be accessed by registered users and portal administrators. The ownership of all data uploaded, stored on the CONP remains with the original owner of the data along with all copyright and license attributes. The data generated within the CONP would belong to the registered users.

The CONP portal was designed to parse existing datasets both externally and within the CONP context. As such, a variety of datasets are available for download within this portal. To access the data, the user will first go to the Data page and filter for data of interest. Clicking on an individual dataset will provide more detailed information and download instructions. There are a number of data download methodologies that will be available to match different user preferences. At the moment, access to datasets is available using the DataLad data management system. One-click data download functionality is under development and will be available in the future.

Once the user has downloaded a dataset, what can they do with it? We also provide a wide range of tools and pipelines. Read the following sections to find out more.

What tool/pipelines are accessible via the CONP Portal?

The CONP portal aggregates a diverse range of 62 tools and pipelines. Many of these tools/pipelines are well-established and have been provided by neuroscience or genomics research institutes. A full list can be found here.

How can I use tools/pipelines?

Several mechanisms are available to run tools and pipelines found on the CONP portal:

  • Tools and pipelines can be run locally on any system using Boutiques and a container engine ( Docker or Singularity ) based on the Zenodo ID of the tool. If you have never used Boutiques in the past, we recommend the following tutorial.
  • Some tools and pipelines can be run on HPCs via the CBRAIN infrastructure.

How can I share/upload a dataset?

The idea of this portal is to have a distributed, automated and simple manner of sharing data. We recognize that having one system to rule them all is not feasible, so our goal is interoperability. Sharing will occur, ideally through API connections for existing platforms, or by providing resources to users without storage/sharing infrastructure.

Data can be uploaded to CONP via the following mechanisms, using:

  • Zenodo by tagging your dataset with the keyword canadian-open-neuroscience-platform
  • The Open Science Framework ( OSF ) by tagging your dataset with canadian-open-neuroscience-platform and setting the dataset to Public
  • Using DataLad within the context of CONP

Detailed technical instructions are available for all these options.