Blue Brain Project

The Blue Brain Project is an attempt to reverse engineer the mammalian brain and to recreate it in a molecular-level computer simulation. The project was founded in May 2005 by Henry Markram of the Brain and Mind Institute (BMI) at the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland. Goals of the project are to enable better and faster development of brain disease treatments, and to satisfy our curiosity about the nature of mind and consciousness.

The research involves studying slices of living brain tissue using microscopes and patch clamp electrodes. Data is collected about every one of the hundreds of different neuron types. This data is used to build biologically realistic models of neurons, and networks of neurons, in the cerebral cortex. The simulations are carried out on a Blue Gene supercomputer provided by IBM. Hence the name Blue Brain. The simulation software is based around Michael Hines's NEURON, together with many other custom-built components.

As of November 2011 the largest simulations were of mesocircuits containing ~100 cortical columns. Such a simulation involves 1 million neurons and 1 billion synapses. This is approximately the same scale as that of a honey bee. It is hoped that a rat brain simulation (100 million neurons) will be achieved by 2014, and a full-scale human brain (100 billion neurons) by 2023.

Contents

Latest news
Background
Objections
Supercomputer
Funding
Collaborators
Timeline
Videos
People involved
Research papers
Weblinks

Latest news

03.Jan.2012 - FET Flagships mid-term conference presentation is now available online: Introducing the Human Brain Project

24.Nov.2011 - HBP presented in Warsaw at the FET Flagships mid-term conference. Presentation will be available online soon.

02.Nov.2011 - Video of a discussion panel held in San Francisco. Markram presents his vision for the Human Brain Project.

Background

One in four people will suffer from one of 560 brain diseases in their lifetime. Therefore it is important to have a good strategy for understanding these diseases and finding suitable treatments. The living brain is very difficult to study. Both from a technical perspective, and a moral one. A virtual model, however, makes direct observations possible. Experiments on models are also more efficient and limit the need for laboratory animals. The Blue Brain Project, being a molecular-level simulation, could be used to study the effect of new pharmaceutical compounds on virtual brains of any species, age, and stage of disease.

Another aim of the Blue Brain Project is to provide a centrally coordinated resource for the 200,000 active neuroscientists in the world. Previously each researcher has focused on their own specialist field without the results being shared and easily available to all. The BBP hopes to build a bigger, better platform for neuroscientists to experiment on. The project is becoming a brain simulation facility that is accessible to all.

There are three main steps to building the virtual brain: data acquisition, simulation, and visualisation.

Data acquisition involves taking brain slices, placing them under a microscope, and measuring the individual neurons with an electrode. This way the different types of neuron are isolated and catalogued. The neurons are typed by morphology, behaviour, and population density. These observations are then translated into mathematical algorithms which describe the neuron's form and function.

The simulation step involves synthesising virtual cells using the algorithms that were found to describe real neurons. The algorthims and parameters are adjusted for the age, species, and disease stage of the animal being simulated. Every single protein is simulated, and there are about a billion of these in one cell. First a network skeleton is built from all the different kinds of synthesised neurons. Then the cells are connected together according to the rules that have been found experimentally. Finally the neurons are functionalised and the simulation brought to life. The patterns of emergent behaviour are viewed with visualisation software.

A basic unit of the cerebral cortex is the cortical column. Each column can be mapped to one function, e.g. in rats one column is devoted to each whisker. A rat cortical column has about 10,000 neurons and is about the size of a pinhead. The latest simulations, as of November 2011, contain about 100 columns, 1 million neurons, and 1 billion synapses. A real life rat has about 100,000 columns in total, and humans have around 2 million. Techniques are being developed for multiscale simulation whereby active parts of the brain are simulated in great detail, quiescent parts not so detailed.

Every two weeks a column model is run. The simulations reproduce observations that are seen in living neurons. Emergent properties are seen that require larger and larger networks. The plan is to build a generalised simulation tool, one that makes it easy to build circuits. There are also plans to couple the brain simulations to avatars living in a virtual environment, and eventually also to robots interacting with the real world. The ultimate aim is to be able to understand and reproduce human consciousness.

Objections

The Blue Brain Project commonly raises objections among both the general public and other neuroscientists.

One objection is that the human brain is simply too complex to simulate. It is argued that there are too many many neurons, and too many supporting glial cells, and each cell contains such an overwhelming number of biochemical processes, that no computer can ever simulate everything. Henry Markram has stated, however, that it is the seeming impossibility of the task that makes it exciting. He says that we shouldn't be discussing whether or not it's possible. Instead we should be asking what it will take to make it possible.

Also, there is no denying that since the 1960s supercomputers have shown a fairly linear increase in their memory capacity and processing power. There's no reason to believe this trend will stop any time in the forseeable future. Given that, it is inevitable that one day supercomputers will be sufficiently powerful to simulate the human brain. We don't yet know for sure when that day will come. Markram claims an exascale supercomputer will suffice, and that such computers will become available by the year 2023.

A second objection is that there is some kind of supernatural spirit at work in the brain, something which is beyond scientific understanding and which is fundamentally impossible to simulate. Markram is a reductionist. He believes that the mind is an emergent property of complex circuits in the brain. He says, however, that if the Blue Brain Project fails then that will be an equally interesting result. It will prove that there is something more going on in the brain that science hasn't yet recognised. Either way, there is only one way to find out for sure, and that's to go ahead and do the experiment.

Supercomputer

agreement between EPFL and IBM which supplied the BlueGene/L supercomputer in 2005(check exact date)
  - later upgraded to BlueGene/P (june 2010)

photos of the computer itself:
http://bluegene.epfl.ch/Presentations/Cadmos_Pres_Clemencon_24sep.pdf

32-processor SGI system for visualisation of results

Uses the Blue Gene/P supercomputer installed on the EPFL campus in Lausanne
  EPFL Wikipedia
Tech specifications: http://www.cadmos.org/fr/caracteristiques-du-blue-gene-p.html
  4096 quad-core nodes (16,384 cores)
  each core is a PowerPC 450, 850 MHz
  Total: 56 TFops, 16 TB of memory
Was 99th fastest November 2009, then 343th in June 2011, has since dropped out of the top 500
  http://i.top500.org/system/176508
  http://en.wikipedia.org/wiki/TOP500
Google Map
Supercomputer useage by week: http://cadmos.epfl.ch/page-60181-en.html

Funding

funding by swiss government
  EPFL is one of the only two federally funded universities in Switzerland
    ...the other being ETH in Zurich
  other grants (what exactly?)
  private individuals (who and how much?)

http://technology-report.com/2009/01/blue-brain-project-ibm-has-not-withdrawn-support/
The BBP project is funded primarily by the Swiss government and secondarily by grants and
some donations from private individuals

The BBP is a project of the Swiss Federal Institute for Technology (EPFL), so funding from
the EPFL (which means from the Swiss government), my research grants (European Union,
Foundations, etc), some other entities and just one special visionary donor.

Edmond J. Safra Philanthropic Foundation headed by his widow, Lily Safra
 - who donated to HU for the establishment of the $130 million Edmond and Lily Safra
  Center for Brain Sciences
 - has given $50 million.
Edmond Safra, a Jewish banker, suffered from Parkinson's disease

http://en.wikipedia.org/wiki/Lily_Safra#Philanthropy_and_art_collection

The Blue Brain Project has applied for an EU flagship grant which would bring in €1 billion in funding over 10 years. If the grant is awarded then the project will be renamed the Human Brain Project. A decision on the funding is due in the second half of 2012 and the project would start at the beginning of 2013. Markram has applied for a €1 billion European grant If awarded, the project will be renamed the Human Brain Project Decision on funding is due in second half of 2012

Collaborators

The Universidad Politécnica de Madrid (UPM) and Instituto Cajal (IC) from Consejo Superior de Investigaciones Científicas (CSIC) are involved in the Blue Brain Project (BBP) with an initiative named Cajal Blue Brain. Different research groups and laboratories from Spanish institutions take part in this initiative, grouping together a large number of scientist, engineers and practitioners.

Idan Segev and team at Hebrew University in Jerusalem, Israel
Phil Goodman of the University of Reno, Nevada
Michael Hines of Yale University, author of NEURON simulator, MP enhancements
Alex Thomson, School of Pharmacy, University of London
Yun Wang, St. Elizabeth's Medical Center, Boston (MA)

Timeline

2002 - Markram founded Brain Mind Institude (BMI) at EPFL
2005 - June, EPFL and IBM agree to launch Blue Brain Project, IBM to install bluegene
       single cellular model
2006 - december, auto-generated cortical column generated, simulated, biologically valid
2007 - November, modeling and simulation of first rat cortical column
2008 - simulations of cortical column construction
       neocortical column (10,000 cells)
       research on determining position and size of functional cortical columns
2009 - June - BlueGene/L replaced by BlueGene/P, doubling of processors
       simulations of cortical construction contine
2010 - December - apply for FP7 grant
2011 - designing of FP7 project
       simulation of multiple columns
       from september 2011 they move into larger dedicated space (3rd floor)
       cellular mesocircuit (100 columns)
2014 - cellular rat brain, 100 mesocircuits
2023 - cellular human brain, 1,000x rat brain -
       needs exascale computing, billion billion Flops, not necessarily at molecular level

Videos

Year Two: August 2011

Year One: January 2010

People involved

Abdeladim Elhamdani	Alejandro S.	Alvaro Martinez
Bruno Magalhaes	Daniel Keller	Deborah La Mendola
Dimitri Christodoulou	Eilif Muller	Emmanuelle Logette
Farhan Tauheed	Felix Schürmann	Georges Khazen
Henry Markram	James King	Jean-Pierre Ghobri
Jesper Ryge	Joe Graham	Joe Graham
Juan Hernando	Julie Meystre	Kamila Markram
Martin Telefont	Maurizio Pezzoli	Melissa Cochrane
Michael Reimann	Monica Favre	Nenad Buncic
Rajnish Ranjan	Ramnath Sagar	Richard Walker
Robert Bishop	Rodrigo Perin	Sean Hill
Sebastien L.	Shruti Muralidhar	Srikanth Ramaswamy
Stefan E.	Thomas T.	Vincent Delattre
Werner Van Geit	Ying Shi