Periodic Reporting for period 4 - GWAS2FUNC (From GWAS to functional studies: Tackling the complex nature of brain disorders)
Période du rapport: 2024-03-01 au 2025-08-31
Brain disorders affect millions of people in Europe and place a heavy burden on families, healthcare systems, and society as a whole. An estimated 179 million Europeans live with conditions such as autism, schizophrenia, depression, insomnia, or dementia. Together, these disorders cost nearly 800 billion euros each year, and this number is expected to rise sharply as the population ages. Despite their impact, most available treatments only reduce symptoms. They do not stop the progression of disease or address the underlying biological causes.
A key challenge is that brain disorders have a complex origin. Many different genetic and environmental factors contribute to risk, and each factor has only a small effect. As a result, two people with the same diagnosis may carry very different combinations of risk factors. Large genetic studies known as genome-wide association studies (GWAS) can detect these risk factors, but the findings are often hard to interpret. Small genetic effects are difficult to translate into biological insights, and even harder to turn into meaningful functional experiments or treatment strategies.
The goal of this ERC project was to close this gap between genetic discovery and biological understanding. In other words, we aimed to explain what the identified genetic risk factors actually mean for the biology of brain disorders. Achieving this required close collaboration between human genetics and neuroscience.
The project focused on three main objectives:
1. Develop and apply new computational methods to combine genetic results with brain-related biological data, allowing clearer interpretation of GWAS findings.
2. Develop and apply new algorithms to understand the genetic differences that exist within and between brain disorders, helping reveal why patients with the same diagnosis may differ at the genetic level.
3. Test whether human stem-cell–based models can be used to examine the functional effects of genetic risk factors, offering a path toward experimental follow-up of GWAS discoveries.
Across the project, we developed several new algorithms, improved widely used genetic analysis tools, and applied these methods to large datasets covering many brain-related traits, with notable advances for Alzheimer’s disease and schizophrenia. We also generated and analyzed human stem-cell–derived neuronal and astrocyte cultures and combined these findings with proteomics and post-mortem brain data. Together, these results provide new insight into the biological pathways affected in brain disorders, including those related to cell-type specificity, local genetic effects, and disease-related changes in neuronal and glial function.
In conclusion, the project successfully advanced the connection between genetic discoveries and brain biology. It delivers computational tools, biological findings, and experimental resources that help move the field closer to understanding the mechanisms behind brain disorders and, ultimately, toward improving treatment strategies.
A key challenge is that brain disorders have a complex origin. Many different genetic and environmental factors contribute to risk, and each factor has only a small effect. As a result, two people with the same diagnosis may carry very different combinations of risk factors. Large genetic studies known as genome-wide association studies (GWAS) can detect these risk factors, but the findings are often hard to interpret. Small genetic effects are difficult to translate into biological insights, and even harder to turn into meaningful functional experiments or treatment strategies.
The goal of this ERC project was to close this gap between genetic discovery and biological understanding. In other words, we aimed to explain what the identified genetic risk factors actually mean for the biology of brain disorders. Achieving this required close collaboration between human genetics and neuroscience.
The project focused on three main objectives:
1. Develop and apply new computational methods to combine genetic results with brain-related biological data, allowing clearer interpretation of GWAS findings.
2. Develop and apply new algorithms to understand the genetic differences that exist within and between brain disorders, helping reveal why patients with the same diagnosis may differ at the genetic level.
3. Test whether human stem-cell–based models can be used to examine the functional effects of genetic risk factors, offering a path toward experimental follow-up of GWAS discoveries.
Across the project, we developed several new algorithms, improved widely used genetic analysis tools, and applied these methods to large datasets covering many brain-related traits, with notable advances for Alzheimer’s disease and schizophrenia. We also generated and analyzed human stem-cell–derived neuronal and astrocyte cultures and combined these findings with proteomics and post-mortem brain data. Together, these results provide new insight into the biological pathways affected in brain disorders, including those related to cell-type specificity, local genetic effects, and disease-related changes in neuronal and glial function.
In conclusion, the project successfully advanced the connection between genetic discoveries and brain biology. It delivers computational tools, biological findings, and experimental resources that help move the field closer to understanding the mechanisms behind brain disorders and, ultimately, toward improving treatment strategies.
From the start of the project, the team worked toward bridging the gap between large-scale genetic discoveries and biological understanding of brain disorders. Work combined computational method development, large consortium-level genetic analyses, and experimental validation using iPSC-derived neuronal systems. Progress unfolded along three lines: new GWAS interpretation algorithms, improved methods for analysing genetic heterogeneity, and functional studies using iPSC models and proteomics.
Key achievements included algorithms that improved biological interpretation of GWAS by identifying trait-relevant cell types, integrating chromatin interaction data, and linking genetic effects to large single-cell RNA-sequencing datasets. These were implemented through major upgrades to FUMA, which gained higher capacity, stability, and more detailed annotation. Harmonised single-cell datasets provided finer biological resolution for many brain-related traits.
The project also developed tools for local genetic correlations and prioritizing causal genes. These tools were applied to numerous brain disorders. Analysis of the largest Alzheimer’s disease GWAS highlighted microglia and neuronal subtypes as key cell types involved in risk. Integration of single-cell and chromatin data refined understanding of how Alzheimer risk is distributed across cell populations. Similar applications clarified shared and distinct genetic pathways across psychiatric disorders including depression, anxiety, schizophrenia, and insomnia.
Functional follow-up work used iPSC-derived neurons and astrocytes from schizophrenia (SCZ) cases and controls, combined with proteomics and multielectrode array recordings. These studies revealed SCZ-associated changes in cellular organisation, synaptic function, extracellular vesicle biology, and neural network activity. Proteomics of post-mortem SCZ cortex showed convergent disruption of mitochondrial and kinase-related pathways, indicating broad metabolic effects. Together, these results demonstrated convergence of genetic risk on specific molecular and functional processes in human neuronal systems.
The project generated high-impact publications, broadly used software tools, and publicly available datasets. Dissemination included journal articles, preprints, conference presentations, training, and integration of new resources into FUMA. Overall, it linked genetic discoveries to biological and functional mechanisms in disorders such as Alzheimer’s disease and schizophrenia and established a foundation for future mechanistic studies.
Key achievements included algorithms that improved biological interpretation of GWAS by identifying trait-relevant cell types, integrating chromatin interaction data, and linking genetic effects to large single-cell RNA-sequencing datasets. These were implemented through major upgrades to FUMA, which gained higher capacity, stability, and more detailed annotation. Harmonised single-cell datasets provided finer biological resolution for many brain-related traits.
The project also developed tools for local genetic correlations and prioritizing causal genes. These tools were applied to numerous brain disorders. Analysis of the largest Alzheimer’s disease GWAS highlighted microglia and neuronal subtypes as key cell types involved in risk. Integration of single-cell and chromatin data refined understanding of how Alzheimer risk is distributed across cell populations. Similar applications clarified shared and distinct genetic pathways across psychiatric disorders including depression, anxiety, schizophrenia, and insomnia.
Functional follow-up work used iPSC-derived neurons and astrocytes from schizophrenia (SCZ) cases and controls, combined with proteomics and multielectrode array recordings. These studies revealed SCZ-associated changes in cellular organisation, synaptic function, extracellular vesicle biology, and neural network activity. Proteomics of post-mortem SCZ cortex showed convergent disruption of mitochondrial and kinase-related pathways, indicating broad metabolic effects. Together, these results demonstrated convergence of genetic risk on specific molecular and functional processes in human neuronal systems.
The project generated high-impact publications, broadly used software tools, and publicly available datasets. Dissemination included journal articles, preprints, conference presentations, training, and integration of new resources into FUMA. Overall, it linked genetic discoveries to biological and functional mechanisms in disorders such as Alzheimer’s disease and schizophrenia and established a foundation for future mechanistic studies.
The project advanced the state of the art in statistical genetics and functional neuroscience. At the start, GWAS interpretation for brain disorders was limited by weak biological annotation, few tools for analysing genetic heterogeneity, and no experimental systems suited to testing effects of small genetic variants. The project directly addressed these limitations.
We introduced new computational methods, including algorithms for cell-type specificity, chromatin interaction integration, and harmonisation of single-cell datasets. These advances, together with major upgrades to FUMA and curated single-cell resources, now allow high-resolution mapping of genetic risk to specific brain cell populations.
LAVA provided a new framework for analysing local genetic correlations and heterogeneity across traits, genomic regions, and molecular phenotypes. Extensions for sex-specific effects, gene–environment interactions, and multi-omics integration represent substantial methodological advances, revealing fine-scale, context-dependent genetic effects in psychiatric and neurological traits.
The project also strengthened functional follow-up of GWAS findings using iPSC-derived neurons and astrocytes from schizophrenia (SCZ) cases and controls.
New methods were completed, implemented, and widely disseminated. Large-scale genetic analyses, including the largest Alzheimer’s disease GWAS, were finalised using these tools. Functional experiments produced disease-relevant signatures and resources that will support future mechanistic work. Together, the computational advances, biological findings, and publicly available datasets ensure lasting impact beyond the project’s duration.
We introduced new computational methods, including algorithms for cell-type specificity, chromatin interaction integration, and harmonisation of single-cell datasets. These advances, together with major upgrades to FUMA and curated single-cell resources, now allow high-resolution mapping of genetic risk to specific brain cell populations.
LAVA provided a new framework for analysing local genetic correlations and heterogeneity across traits, genomic regions, and molecular phenotypes. Extensions for sex-specific effects, gene–environment interactions, and multi-omics integration represent substantial methodological advances, revealing fine-scale, context-dependent genetic effects in psychiatric and neurological traits.
The project also strengthened functional follow-up of GWAS findings using iPSC-derived neurons and astrocytes from schizophrenia (SCZ) cases and controls.
New methods were completed, implemented, and widely disseminated. Large-scale genetic analyses, including the largest Alzheimer’s disease GWAS, were finalised using these tools. Functional experiments produced disease-relevant signatures and resources that will support future mechanistic work. Together, the computational advances, biological findings, and publicly available datasets ensure lasting impact beyond the project’s duration.