Platforms Overview

The initial versions of the Platforms were released on 30 March 2016. This was the result of an extensive multidisciplinary effort involving more than 750 scientific collaborators and engineers from over 100 institutions in 24 countries.

The Platforms are as follows:

  • The Neuroinformatics Platform
  • The Brain Simulation Platform
  • The High Performance Analytics and Computing Platform
  • The Medical Informatics Platform
  • The Neuromorphic Computing Platform
  • The Neurorobotics Platform

HBP Collaboratory and the HBP Platforms

Where collaboration happens

HBP Collaboratory

The HBP Collaboratory collects tools from the HBP Platforms in one place and allows you to organize them into your own collaborative workspace or collab.

 

  • Create a collab, a collaborative scientific workspace
  • Build your team
  • Describe your project for the collab team and the world
  • Collect and organize the tools you want to use
  • Share data
  • Use Jupyter notebooks to share your ideas, code and workflows

A collage of Collaboratory screenshots

 

 

Neuroinformatics Platform

Neuroinformatics platform organizes neuroscience data from the HBP and beyond.

The goals of Neuroinformatics Platform is to provide to the neuroscience community with:

  • An integrated multi-level, data enriched atlas of the rodent brain.
  • An integrated multi-level, data enriched atlas of the human brain.
  • Tools to curate and spatially register datasets to standard reference atlases.
  • Identify or develop ontologies to qualify and describe data, enabling data navigation and search.
  • Analytics pipeline to automatically process sub-sets of data in order to provide and integrated view of the multi-level atlases to users.
  • 2D and 3D Viewers to navigate, query and annotate the atlases.
  • Programmatic API to access data in the atlases.

Learn more about the HBP Neuroinformatics Platform

Human Atlas with grey matter / white matter parcellation

 

Brain Simulation Platform

This is the first public prototype of the Brain Simulation Platform. It entails functionality spearheaded in the first 30 months of the HBP, of a total 10 year development cycle. Accordingly, the functionality at this point has different levels of maturity and public accessibility. In brief, many of the fundamental software packages that the Platform relies on and develops, are freely available as open source.Neuroinformatics platform organizes neuroscience data from the HBP and beyond.

Intricate pipelines for reconstruction and simulation are packaged into web-accessible workflows, showcased as use cases, typically requiring privileged access. Lastly, using these tools and workflows, the Platform hosts team-science Collabs for collaborative building of scaffold models of different brain regions, serving as co-design drivers for the Platform.

Access the HBP Brain Simulation Platform

A portion of the Hippocampus model building Jupyter notebook workflow

 

High Performance Analytics and Computing Platform

The High Performance Analytics and Computing (HPAC) Platform provides neuroscientists with the high performance computing (HPC), storage and data processing capabilities they need to run simulation of sophisticated, detailed brain models and to analyse large, complex data sets. It also provides software tools and frameworks for scientific visualisation, interactive and visual data analytics, performance analysis of parallel applications, data management, time-series analysis, dynamic load balancing, parallel programming models, and numerical models for brain simulations.
 
Some HPAC Platform users need computing resources themselves; others also would like to access datasets and results produced and stored at the HPC centres. The HPAC Platform offers software tools developed by the Platform partners for both groups of users. The execution of some of these tools, for example for visualization and data management, does not necessarily require a supercomputer, but they can be used on standard computers and notebooks. Other software, like parallel programming frameworks, can be used on both types of architectures. Neuroscientists who would like to use a supercomputer need experience in programming with languages like C, C++, Python or Fortran. First experience with HPC is very helpful, but the HPC centres also offer introductory courses and support.
 

Learn more about the HBP HPAC Platform

A picture of a part of the Juelich datacenter under construction

 

Medical Informatics Platform

The aim of the Medical Informatics Platform is to provide researchers the ability to access and analyse large amounts of anonymised clinical data from hospital, research, and pharmaceutical clinical trial databases through an innovative data management system that we are building.

The system integrates heterogeneous data formats seamlessly and federates data sources into a harmonized virtual database with a customized interface for navigation and data mining. The patterns discovered in the data ("biological signatures" which uniquely identify diseases) will generate new hypotheses about brain diseases for investigation and will lead to their novel classification, the latter based on biological, physiological and anatomical features in addition to the classical patterns of phenomenology expressed in symptoms, signs and syndromes.

The data will also be available to answer public health and medical epidemiological questions proposed by the community of medical scientists and planners. In the long run our vision is that unlocking the wealth of information locked up in medical and research databases will provide a credible and rapid path to precision (or personalized) medical care.

Our interaction with the computing and neuroscience components of the HBP will serve to test with lesion models the functional and structural brain models and artifacts they produce.

Learn more about HBP Medical Informatics Platform

Medical Informatics Platfrom screenshot

 

Neuromorphic Computing Platform

The Neuromorphic Computing Platform developed in the Human Brain Project (HBP) provides remote access to two complementary, large-scale neuromorphic computing systems (NCS) built in custom hardware at locations in Heidelberg (the BrainScaleS system) and Manchester (the SpiNNaker system).

The NCS are programmable, brain-inspired computing devices which enable high-speed, low-energy simulations of spiking neural networks with synaptic plasticity.

The BrainScaleS system is based on physical (analogue or mixed-signal) emulations of neuron, synapse and plasticity models with digital connectivity, running up to ten thousand times faster than real time.

The SpiNNaker system is based on numerical models running in real time on custom digital multicore chips using the ARM architecture.

 

Models and simulation experiments are described in a Python script using the PyNN API. Experiments can be submitted in your browser through the HBP Collaboratory or via our web API (Python client available). Similarly, once the simulation is complete you can view the results in your browser, download data files via the web API, or transfer the data to the HBP High Performance Computing platform for further analysis.

Learn more about the HBP Neuromorphic Computing Platform

A neuromorphic computing platform collage

 

Neurorobotics Platform

The NeuroRobotics Platform (NRP) is an Internet-accessible simulation system that allows the simulation of robots controlled by spiking neural networks.

You access the platform from your browser (see list of supported browsers). When you log into the platform you can start a new experiment or join other experiments that are currently running. If you start a new experiment, you can change the experiment (currently just some parameters) and interact with the simulation. If you join a running experiment, you are only an observer, but you can still navigate through the scene.

The platform should be easily accessible to researchers from neurosciences and robotics alike with little to no programming involved for setting up and running experiments.

Learn more about the HBP Neurorobotics Platform

A neurorobotics platform collage

 

Enabling new kinds of collaborative research

The Platforms will enable new kinds of collaborative research to be performed in brain research, cognitive neuroscience and brain-inspired computing. The prototype tools, hardware systems, and initial datasets are designed to enable faster and more efficient research techniques in, for example, modelling, in silico experimentation, or data analysis. Users are encouraged to explore the Platforms and build interactive "collabs" as part of the HBP Collaboratory, a web-based scientific research hub and the main entry point to the Platforms.

The Platforms will be further developed based on input and feedback from the users; the functionalities of the Platforms will therefore expand and evolve over the course of the Project.

Accessing the HBP Platforms – the HBP Collaboratory

The HBP Collaboratory is the main entry point to the Platforms. The HBP Collaboratory and the HBP Platforms are subject to restrictions on their use. In most cases these restrictions are due to limited computing or storage capacity powering the Platform service offerings. More information is available here.

There are six user categories; each of which gives a different type of access to the Platform features and tools. The user categories are as follows:

  1. Public
  2. HBP Identity Account
  3. HBP Platform Test Access
  4. HBP Platform Project Access
  5. HBP Partnering Projects
  6. HBP Member Account

For information on the Platform features each user category can access, please visit the User Access page.

Terms of service and data protection

The HBP is dedicated to responsible and ethical research. It recognises the need to comply with European and national regulation and legislation. This explicitly includes matters of data privacy, data protection and ethical issues related to data generation and use (see here for an overview). Data protection has been a focus of the HBP Ethics and Society Subproject, and the result of their work and the collaboration with the other Subprojects has informed the way in which the HBP deals with such matters.

At present, each of the Platforms has a set of terms and conditions, which include a section on data. Users of each Platform will need to accept these terms in order to access the Platform. These terms and conditions determine which activities users can engage in.

For further questions concerning ethical issues and how these relate to the data governance of the HBP, please contact the HBP Ethics Manager, Professor Bernd Carsten Stahl.

Working together

The HBP Platforms are being developed with input from end users. Some components, such as the Neuromorphic Computing Systems that underpin the Neuromorphic Computing Platform, also build on tools developed as part of earlier initiatives such as SpiNNaker, BrainScaleS, FACETS, NEST and NEURON.

The HBP has made a significant contribution to further developing these existing tools, creating many new tools and, importantly, in bringing them together and making them accessible to scientific users through a single entry point, the HBP Collaboratory.

Contact us 

We value your feedback.

For general questions and feedback, please contact us at info@humanbrainproject.eu.

For technical questions, please contact us at platform@humanbrainproject.eu.