Brain disorders, including neurodegenerative diseases and mental illnesses, are often difficult to diagnose, hampering fundamental research and drug development. This complexity is caused by clinical and neuropathological heterogeneity, overlap in clinical manifestations between disorders, and frequent comorbidities. Hence, there is a clear need for data-driven approaches to disentangle these complex disorders for improved diagnosis and prognosis.
We recently established the Netherlands Neurogenetics Database, which takes a cross-disorder approach to study the Netherlands Brain Bank, one of the largest and best characterized brain autopsy cohorts in the world. For this we have implemented extensive sets of parsers and Natural Language Processing techniques to convert 1) medical records summaries to clinical disease trajectories, 2) neuropathological reports to neuropathological trait assessments. We also have obtained genotype information for almost 3000 donors that allow us to calculate polygenic risk scores for brain disorders, and neuropsychological traits. These different data types are used for dimensionality reduction approaches, to identify molecular and clinical endophenotypes, and for predictive modelling to predict neuropathological diagnosis.
To model key findings, we are also involved in establishing brain organoid models for neurological and psychiatric disorders.
Brain autopsy studies show that many patients with brain disorders were clinically misdiagnosed. Moreover, after a patient receives a particular diagnosis, it is very difficult to predict how the disease will develop. Hence, there is a strong need for improved diagnosis and prognosis. The Netherlands Brain Bank has one of the largest and best characterized brain autopsy cohorts in the world. By studying the symptomatology of these donors, in conjunction with neuropathology, and genetics, my research group aims to build predictive models and statistical tools to better delineate and predict brain diseases and disease progression.
A key aim of the lab is to identify transcription factors that drive the pathogenesis of brain disorders, including transcription factors that drive microglia subsets in Alzheimer’s Disease, and transcription factors tjat drive neuronal differentiation patterns in mental illnesses.
Genome Wide Association Studies (GWAS) have identified hundreds of loci in the genome that are associated with brain disorders, including mental illnesses, multiple sclerosis and Alzheimer’s Disease. However, the mechanisms by which these risk loci exert their effect is currently largely unknown. We aim to study genetic susceptibility by integrating state of the art computational tools with multi-omics data types. For this, we focus on individual disease risk loci, but also on polygenic risk scores. We also aim to functionally test these predictions using brain organoids in collaboration with a post-doc, Marina Trombetta Lima, that I co-supervise.
Over the last few years, we have implemented extensive sets of parsers and Natural Language Processing (NLP) techniques, including Google Bert and T5, to convert extensive clinical/pathological summaries from the Netherlands Brain Bank to standardized trait assessments, and clinical disease trajectories.
University Medical Center Groningen (UMCG)
Department of Biomedical Sciences of Cells and Systems
Section Molecular Neurobiology
Internal Zipcode FB43
Antonius Deusinglaan 1
9700 AD Groningen
The Netherlands
Visiting address
University Medical Center Groningen (UMCG)
Department of Biomedical Sciences of Cells and Systems
Section Molecular Neurobiology
Antonius Deusinglaan 1
9713 AV Groningen
The Netherlands
location: Building 3215, 8th floor
