Development of molecular brain atlas with machine learning

Development of molecular brain atlas with machine learning
Development of molecular brain atlas with machine learning

ENGINEERINGNET.BE – One in four people suffer from a brain disorder, such as depression or frontotemporal dementia.

“These disorders may be related to changes that occur in individual cells, and in particular between glial cells and nerve cells,” says Professor Maarten Kole of the Netherlands Brain Institute and UUtrecht.

“More and more brain material from people with psychiatric disorders is available to investigate this issue. Promising techniques have also been developed in chemistry and molecular biology that can change research into the human brain.”

Through this study, the researchers hope to gain more clarity about what exactly happens in the cells and to bring this together into a molecular brain atlas. Various disciplines, such as neurobiology, chemistry, data science and psychiatry, join forces to this end.

Kole: “The goal is to develop biomarkers for the diagnosis, prognosis and treatment of brain disorders. Once we have a better idea of ​​what exactly is going wrong in the brain, we will soon be able to take targeted measurements in the blood or with PET scans to determine what is going on. This allows us to move towards a more personalized approach.”

The researchers take human brain material from the Dutch Brain Bank as a starting point. They also collect brain fluid and plasma from living patients with a brain disorder.

The molecular brain atlas is created by mapping in great detail the changes that take place in sixteen brain areas that are known to play a role in neuropsychiatric disorders.

Using machine learning models, the researchers want to link the large amounts of data that are collected to the psychiatric symptoms of around 3,000 donors.

The researchers can then use innovative chemical methods to investigate in model systems, such as mini-organs of brain tissue grown in the lab, how changes linked to certain symptoms can be influenced or corrected.

The research is led by the Institute for Chemical Neuroscience. Partners in the project are various Dutch universities, university medical centers and the Netherlands Brain Institute. Funding comes from the Netherlands Organization for Scientific Research.

The article is in Dutch

Tags: Development molecular brain atlas machine learning


PREV New research advises against vaping near children: “Link with cancer”
NEXT What type of magnesium do you need?