Blue Brain Nexus: an open-source knowledge graph for data-driven science

EPFL’s Blue Brain Project creates and open sources Blue Brain Nexus, which allows the building of data integration platforms. Blue Brain Nexus enables data-driven science through searching, integrating and tracking large-scale data and models.

Blue Brain Project deploys HPE supercomputer for digital reconstruction and simulations of the mammalian brain to advance the understanding of the brain

In July, Hewlett Packard Enterprise (HPE) announced that the EPFL Blue Brain Project had selected HPE to build a next-generation supercomputer for modeling and simulation of the mammalian brain. The new supercomputer, called ‘Blue Brain 5’, will be dedicated to simulation neuroscience, in particular simulation-based research, analysis and visualization, to advance the understanding of the brain.

The Blue Brain Portal – a knowledge space for simulation neuroscience

Released in August, the Blue Brain Portal brings together in one place open-sourced software, tools, models and data, both from us and our collaborators. The aim is for this knowledge to be utilized by both the neuroscientific and the wider scientific community to develop the field of simulation neuroscience.

Blue Brain Project releases first-ever digital 3D brain cell atlas

Like “going from hand-drawn maps to Google Earth,” the Blue Brain Cell Atlas allows anyone to visualize every region in the mouse brain, cell-by-cell – and freely download data for new analyses and modelling.

The first digital 3D atlas of every cell in the mouse brain provides neuroscientists with previously unavailable information on major cell types, numbers and positions in all 737 brain regions – which will potentially accelerate progress in brain science massively. Released by EPFL’s Blue Brain Project and published in Frontiers in Neuroinformatics, the Blue Brain Cell Atlas integrates data from thousands of whole brain tissue stains into a comprehensive, interactive and dynamic online resource that can continuously be updated with new findings. This groundbreaking digital atlas can be used for analyzing and further modeling specific brain areas, and is a major step toward a full simulation of the rodent brain.


Blue Brain Team Discovers a Multi-Dimensional Universe in Brain Networks

A team of scientists led by the Blue Brain Project used a sophisticated type of mathematics in a way that it has never been used before in neuroscience. The team uncovered a universe of multi-dimensional geometrical structures and spaces within the networks of the brain.

This research, published in Frontiers in Computational Neuroscience, has significant implications for our understanding of the brain.

Blue Brain Project launches three-day conference to kick-start neuromodulation research – NM2

The NM2 Conference was conceived to address understanding the mechanisms by which neuromodulators operate which is both fundamental to Blue Brain’s pioneering work in simulating brain function and dysfunction, and for the global neuroscience community

Leading experts from around the world and EPFL presented and took part in panel discussions across the three days. Additionally, the NM2 Conference provided a unique platform for students and junior researchers to interact with leaders in the field to collectively take part in shaping the future course of neuromodulatory research.


The Blue Brain Project releases the Blue Brain Python Optimization Library (BluePyOpt) – an extensible open source framework for data-driven model parameter optimisation that wraps and standardises several existing open-source tools. The library includes methods for setting up small- and large-scale optimizations on a broad range of compute platforms – from laptops to large cloud-based compute infrastructures.


Blue Brain reaches a major milestone with the publication of a first draft of the digital reconstruction of neocortical microcircuitry (Markram et al, 2015). The study confirmed the feasibility of building and simulating a digital copy of a part of the brain and demonstrated that multidisciplinary Big Science in the field of neuroscience yields favorable results (82 scientists contributed to the study).

The paper, which appeared in the journal Cell, represents the most complete description of any neural microcircuit to date. It provides a complete digital map of all the cells and synapses in a block of neural tissue and describes simulation experiments replicating a range of previous in vivo experiments. In other words, our digital copy of a part of the brain behaves like a real part of the brain.

Most significantly, this study advances the case for simulation as a useful new method in neuroscience. It proves that we understand the basic properties of the components and interconnections of the brain well enough to be able to reconstruct and simulate certain physiological functions. This advance makes it feasible, in principle, to reconstruct the human brain even though we can never measure all its parts.


The BBP computing team works to improve the efficiency and scope of BBP computing tools and supercomputing infrastructure. A series of publications describe the new tools.

In June, the BBP replaces its previous supercomputer (the BlueGene/P) with a BlueGene/Q machine (Blue Brain 4) hosted at the Swiss National Computing Centre (CSCS) in Lugano. The new machine offers higher performance and expanded memory.

In the same month, BBP, IBM Research and ETH Zürich announce a collaboration to develop a new hybrid memory strategy for supercomputers, matching the heavy memory requirements for reconstructions of large volumes of neural tissue (brain regions, whole brains).

The BBP completes validated digital reconstructions of neural microcircuitry in the brain of young rats. Work begins on a major paper, presenting the reconstruction, and on online resources, making the results available to the broader community.


On January 28, the EU Commission announces that it has selected the Human Brain Project as one of its two FET Flagship projects (cordis.europa.eu/fp7/ict/programme/fet/flagship/doc/press28jan13-01_en.pdf). Work on the project begins in October 2013, with the Blue Brain team playing a leading role. EPFL hosts the project’s first “Summit Meeting”

The Blue Brain Project is officially granted the status of a Swiss National Research Infrastructure, funded by the ETH Board

Two important BBP publications describe the use of BBP models ed to identify and characterize “neuronal clusters” in neural microcircuits, and to predict local field potentials.


In April, Human Brain Project Consortium concludes its preparatory study and publishes a public report. In October the HBP consortium submits its formal application to become a FET Flagship project. The Blue Brain team coordinates the preparation of the proposal.

An important paper in PNAS describes BBP-developed methods, making it possible to predict the connectivity of neocortical microcircuitry.

At the Neuroscience 2012 conference in New Orleans, the Blue Brain Project presents more than 20 posters, describing a first reconstruction of the rat cortical column.


In January, the European Commission informs the Human Brain Project consortium that it has been selected to perform a preparatory study. Work on the study begins in May, coordinated by members of BBP.

The project hires new engineers and scientist. In November the enlarged team moves to new office space in the EPFL Innovation Park. The project publishes several high impact papers describing new methods to generate cell models and in silico studies of virtual brain tissue.


The BBP drives the formation of a Consortium to participate in the European Commission’s newly launched FET Flagship Programme. In December, the new consortium applies to the Commission (Seventh Framework Programme – FP7) to fund a large scale research project – the Human Brain Project. The goal of the new project is to understand the human brain and its diseases and, ultimately, to emulate its computational capabilities. A key objective is to reconstruct and simulate the whole human brain. The approach described in the project proposal builds on the methods and tools developed in the BBP.


In June, thanks to the CADMOS initiative, the BBP’s BlueGene/L supercomputer is replaced by a BlueGene/P, with double the number of processors . The new machine represents a major increase in BBP computing power.

“In silico” experimentation is in full swing, testing the behaviour of BBP models against results from other research groups. The results provide new insights into the principles underlying the construction of neocortical microcircuitry.


The BBP team tests the accuracy of its model-building against anatomical and physiological data from laboratory experiments. In June, an article in the HFSP Journal summarizes the on-going debate on the size and location of functional cortical columns


In January, Henry Markram presents the project to the Davos forum. November 26 marks the end of the first phase of the project, which announces the completion of an initial model of the rat cortical column


In February, the project takes shape. An article in Nature Reviews Neuroscience by Henry Markram describes the project’s goals and methods. During the summer, the BBP team generates its first model of a cortical column, using a simplified neuron model..


In June, the EPFL and IBM sign an agreement to launch the Blue Brain project (BBP). The agreement provides for the installation of a BlueGene supercomputer on the EPFL campus.