Daniel Manrique-Castano, Neuroscientist, Germany

16 April 2020

This article was originally published in Spanish at https://hablemosdeneurociencia.com/simular-cerebro-reto-del-human-brain-project/

In October 2013, the European Commission for the Future and Emerging Technologies (FET) launched an international research initiative with an ambitious goal: to simulate the human brain. With this project, the Human Brain Project (HBP), the European Union established a state-of-the-art research platform that will study the brain interdisciplinarily for 10 years, from molecular scales to higher cognitive processes.

To achieve this goal, the research partnerships involved set out six paths to follow:
1. Create an infrastructure in the European Union for brain research.
2. To study and disseminate data to understand diseases of the nervous system.
3. To develop theories and models that allow us to understand the functioning of the brain
4. Simulate the brain
5. Develop brain-inspired computing, data analysis and robotics,
6. Ensure that all this international effort benefits society.

For those of us involved in neuroscience research these goals are ambitious and difficult to achieve, but at the same time, they are a fascinating challenge. While the task is not easy, it is an incentive to do the best science we can.

Will the Human Brain Project help us understand the brain?

First of all, the HBP has developed high-resolution three-dimensional atlases of the brain that allow us to elucidate the cytoarchitecture of the nervous tissue with the most recent anatomical classification criteria [2]. The first, called the BigBrain Atlas, is an atlas of the human brain available as an online platform, which was developed thanks to the collaboration between Canadian and German researchers. The second, the Waxholm Atlas, is an open access volumetric atlas that provides the anatomy of the Sprague Dawley rat’s brain. This resource was constructed from magnetic resonance imaging, which allows observation of the cytoarchitecture of the tissue; and diffusion tensor imaging (DTI), which shows the connections between different areas of the brain.

Secondly, some progress has been made in understanding different cognitive processes and central nervous system (CNS) pathologies. For example, the metabolic processes of patients with altered states of consciousness or coma have been studied to aid diagnosis and clinical treatment [1]. Transcranial magnetic stimulation has been identified to activate the neuronal dendrites in the upper cortical layers, allowing non-invasive treatment of the CNS [5]; and changes in the synaptic activity of individual neurons during different phases of sleep have been explored for the first time [6]. The project is currently focused on the area of cognitive sciences, exploring episodic memory, multisensory recognition and reconstruction, and the mechanisms of consciousness.

The ambition to simulate the brain rest on mathematical principles and the testing of theories around the functioning of the CNS. Similarly, with sufficient simulation tools, one can reduce the number of animal experiments, validate experimental data from in vivo models with computer models, and study CNS diseases in silico. To this end, the HBP developed the Brain Simulation Platform (BSP), a compendium of computational tools that allow scientists to reconstruct and simulate brain systems at different levels, from molecular and cellular models to complex cognitive processes and psychiatric pathologies. Although it may sound strange, I myself have witnessed the work of the Department of Computational Neuroscience at Ruhr Universität Bochum in Germany, where MathLab is used to simulate the different phases of depression.

“I myself have witnessed the work of the Department of Computational Neuroscience at Ruhr Universität Bochum in Germany, where MathLab is used to simulate the different phases of depression.”

Therapeutic effects are also introduced, mathematically, to evaluate the possibilities of recovery or relapse. A rather strange approach for my clinical profile, but at the same time, a path that points towards the logical-mathematical principles of the functioning of the CNS. The project also developed the Neuromorphic Computing Platform, which uses aspects of CNS biology to be digitally copied into electronic circuits. The aim is to develop a tool to understand the processes of human learning and development, using machine learning platforms and, on the other hand, to draw inspiration from the neuronal model to generate computation.

There is a main core called SpiNNaker, located in the city of Manchester, England, which connects 500,000 processors in a network for the exchange of neuronal signals. In turn, the BrainScaleS is a device located in Heidelberg, Germany, which implements analog electronics of 4 million neurons and one billion synapses in silico. As a result of this high computational power, it is expected that in the coming years image and sound recognition algorithms can be developed for devices such as personal computers or cell phones. Later on, in a long term perspective, these computing technologies will allow the integration of electronic devices that help with domestic tasks. In other words, if everything goes as expected, these two machines will be the precursors of the first commercial robots in the near future. 10, 20, 50, 100 years? It’s hard to say, but it will happen at some point. 

SpiNNaker core machine located in Manchester. 

Pabogdan / CC BY-SA (https://creativecommons.org/licenses/by-sa/4.0)y

On-line medical platform

One of the most outstanding contributions of the HBP is an online platform that allows interaction between the different stakeholders in the health area, researchers, epidemiologists, clinicians, developers and pharmaceutical companies. The main objective of this platform is to make effective use of data available worldwide. In this way, the Medical Informatics Platform (MIP) aims to provide the infrastructure and tools to encourage collaborations to determine the different biological mechanisms that lead to CNS pathologies. The first version of the platform was launched in 2016 for the use of European hospitals and research communities that were recruited to make a first evaluation of the project. Researchers who wish to join the network can apply for access to the platform on the HBP website [3].

Through diffusion tensor imaging (DTI) it can be appreciated how different areas of the brain are connected

Public Education

The HBP’s disclosure intentions should also be highlighted. There are different scenarios in which science and medicine generate curiosity, and in some cases, concern in different communities. For example, at an event I attended last week, Carlos Moedas, the European Union’s Commissioner for Research, Science and Innovation, expressed his concern about the public’s knowledge of science. He said that in France 40% of people do not believe in the benefits of vaccines, but they also do not know how a vaccine works.


The commissioner said that one of the most important roles of science in the 21st century is to show the public how the research process works and the benefits it brings to different aspects of life. Fortunately, the HBP takes that into account and among its objectives is to create spaces for outreach and dialogue with the public to resolve emerging controversies. To this end, for example, the Danish Board of Technology has organized meetings and seminars to discuss0 the HBP, robotics and computing.

Only until 2023 will we know what the scope of the BPH was. The simulation of the functional principles of the brain is an arduous task, but it is a task in progress. Every day I see how my office colleague, Egor Dzyubenko, tries to incorporate more and more variables into his neural network simulations. Now, he is trying to add astrocytes to the network, a very important variable [4]. It is likely that today’s generation of young adults will witness the first domestic robots, in every sense of the word. How much can we learn about the brain? How far will the computing power go? Will we reach any insurmountable limit? Let us hope that human cleverness will take us further and further.

References

[1] Bodart, O., Gosseries, O., Wannez, S., Thibaut, A., Annen, J., Boly, M., … Laureys, S. (2017). Measures of metabolism and complexity in the brain of patients with disorders of consciousness. NeuroImage: Clinical, 14, 354–362.

[2] Human Brain Project (2017). Explore the Brain.

[3] Human Brain Project (2017). Medical Informatics Plataform.

[4] Manrique-Castaño, D. (2016). El cerebro es mucho más que neuronas: La unidad Neurovascular.

[5]. Murphy, S. C., Palmer, L. M., Nyffeler, T., Müri, R. M., & Larkum, M. E. (2016). Transcranial magnetic stimulation (TMS) inhibits cortical dendrites. ELife, 

[6] Olcese, U., Bos, J. J., Vinck, M., Lankelma, J. V., van Mourik-Donga, L. B., Schlumm, F., & Pennartz, C. M. A. (2016). Spike-Based Functional Connectivity in Cerebral Cortex and Hippocampus: Loss of Global Connectivity Is Coupled to Preservation of Local Connectivity During Non-REM Sleep. Journal of Neuroscience, 36(29), 7676–7692.

Want to know what’s new in ScienceLogs?

en_USEnglish

Pin It on Pinterest

Share This