3D-BrAIn: Developing a bio-digital twin model of the human brain
The EIC Pathfinder Open project ‘3D-BrAIn’ aims to revolutionize personalized precision medicine for central nervous system (CNS) disorders through development of a bio-digital twin model of the human brain. The project combines breakthrough human brain modelling technology, state-of-the-art 3D multi-electrode array technology, and a novel data analytical approach using tailored biology-inspired automated artificial intelligence (AI)-based algorithms. Erasmus Medical Center (NL) lead the project, while partners 3Brain (CH), Ludwig-Maximilians-University (DE), and the University of Genoa (IT) complete the consortium. 3D-BrAIn was awarded €2 million under the EIC Pathfinder Open call in May 2022 and commenced in April 2023. We spoke to Project Leader, Dr Femke de Vrij, Associate Professor of Stem Cell Modelling at the Department of Psychiatry of Erasmus Medical Center. Femke shared insights into the project, future impact of 3D-BrAIn, and the collaboration with Catalyze towards the successful submission.
Studying neuropsychiatric disorders
Femke’s research background is in modelling brain diseases and investigating their underlying biology. In particular, her research utilizes human stem cell modelling to study neuropsychiatric disorders – a field of research shrouded in obscurity to researchers due to the complexity and inaccessibility in studying the brain.
Femke begins, “Our research focuses on the frontal cortex of the brain, where we think the higher cognitive functions – implicated in psychiatric disorders – are mostly situated”.
“We know a lot from mouse models and we know a lot from imaging, but at a cellular level it remains difficult to really look at human-specific biology, which you could argue is where we should focus for neuropsychiatric disorders.”
Developing a highly reproducible model of the human brain
Currently, clinical management of CNS disorders relies largely on trial and error, while drug development for CNS diseases has been notoriously unsuccessful, with few innovative new drugs being approved in the past decades. The development of a precision medicine platform – a bio-digital twin model of the human brain – to model CNS disorders has transformative potential, for both CNS research and treatment of patients of CNS disease.
While new technologies in genetics and cellular models have furthered research of CNS disorders, it remains difficult to study subtle differences in phenotypes present in psychiatric disorders. This is chiefly what 3D-BrAIn aims to address, by developing a reproducible, homogenous model that can enable more rigorous study of the human brain.
Femke explains, “Simple 2D stem cell models of brain cells in culture tell you a lot, but they don’t form structures. And if you use 3D whole brain organoids – that form organized structures – they can be useful, but they are also very heterogeneous and hard to study.”
“We have developed an intermediate between 2D models and whole brain organoids, where we generate 3D adherent cortical organoids derived from patient-induced pluripotent stem cells (iPSC). These cells self-organize and develop a radial structure, similar to real brain development, where we can distinguish different subtypes of neurons and glial cells, and also different cortical layers. It’s not perfect, but it is very reproducible, which is crucial.”
Studying electric synaptic connectivity of neurons
As Femke explains, their strongest interest is in studying the electric synaptic connectivity of neurons. For this, they use multi-electrode arrays (MEA) – a technology that allows researchers to record spontaneous firing activity of cultured neurons over a period of weeks or months. Disruption of the synaptic connectivity of neurons is implicated as a causal factor for neuropsychiatric diseases.
Femke lays-out how the 3D-BrAIn consortium can make this possible:
“Currently, there are no MEA methods that fit with our new model. 3Brain make such multi electrode arrays at a very high density, and they can have them custom made for our organoids. Also in the project are a neurodevelopmental group from the Ludwig-Maximilians-University , that has a lot of expertise in organoids in different diseases that are well suited to study in this model. Then there’s the LISCOMPlab group at the University of Genoa, that can employ AI solutions to analyze the enormous amounts of data that we will get from these organoids.
“The project is basically aiming to develop a multidisciplinary tool to allow robust and accurate modelling of the CNS across a broad range of neuropsychiatric diseases, including epilepsy, autism, schizophrenia, and Dravet Syndrome.”
Inception of the project, and enlisting the support of Catalyze
Catalyze provided support during the preparations of the project application to the Pathfinder Open call. Femke describes the decision to work with Catalyze, and the added value of this support.
“We saw this EIC Pathfinder Open call and thought it seemed very suitable, but the deadline was very close. I think in particular for the EU grants, there is so much administration in the portal and with many extra documents required. It was really nice to have Catalyze’s support who could take this out of our hands, enabling us to focus on the scientific content.
“In particular, our consultant had very good writing skills and poured all of the content into a great format that we could nicely work together on. I think Catalyze’s help was instrumental in getting it all in shape. We are continuing our collaboration with Catalyze as they will now perform project management for the duration of the 3D-BrAIn project.”
Start-to-end solution for personalized medicine, drug screening and neurotoxicity testing
The 3D-BrAIn technology delivers three-fold capabilities: personalized – i.e. based on patient-derived iPSCs; precision – i.e. advanced computational solutions for innovative patient stratification and drug discovery; and prediction – i.e. individual patient response efficacy to specific medications.
“We hope that this model will be standardized enough that you can essentially get an electrical signature of cortical organoids per individual or patient. Then with artificial intelligence, you can start comparing that in an unbiased way. Beyond personalized use of the 3D-BrAIn model, it can also be used more broadly to stratify patients with a certain disease into specific disease phenotypes.”
Femke goes on to explain the long term ambitions for the bio-digital 3D-BrAIn platform:
“In the long-term we want to use this model to study the neurons in the organoids at single cell resolution, to look at diseases like schizophrenia, where we are interested in certain genes that we can model specifically with 3D-BrAIn. We will try to see if this model will have a good enough validity to tease apart phenotypes or treatments.”
Catalyze is proud to have supported this successful submission to the prestigious EIC Pathfinder Open and thrilled to continue our support through project management.
- Learn more about the research at Erasmuc MC Neurobiology of Mood & Psychotic Disorders group
- Learn more about the EIC Pathfinder programme.
- NEWS: Catalyze supports two successful applicants in 2022 EIC Pathfinder Open