‘Rare genetic variants confer a high risk of ADHD and implicate neuronal biology‘ has been published in Nature. Our team contributed to this study by generating and analyzing protein-protein interaction networks in iPSC-derived neurons to study three rare variant ADHD genes (MAP1A, ANO8, ANK2). Read more about the study in the press release from Aarhus University.

‘A foundational neuronal protein network model unifying multimodal genetic, transcriptional, and proteomic perturbations in schizophrenia‘ is now available on medRxiv.

Abstract:

Schizophrenia (SCZ) is a complex psychiatric disorder with a diverse genetic landscape, involving common regulatory variants, rare protein-coding mutations, structural genomic rearrangements, and transcriptional dysregulation. A critical challenge in developing rationally designed therapeutics is understanding how these various factors converge to disrupt cellular networks in the human brain, ultimately contributing to SCZ. Towards this aim, we generated multimodal data, including SCZ-specific protein-protein interactions in stem-cell-derived neuronal models and adult postmortem cortex, integrated with genetic and transcriptomic datasets from individuals with psychiatric disorders. We identified three distinct neuron-specific SCZ protein networks, or modules, significantly enriched for genetic and transcriptional perturbations associated with SCZ. The relevance of these modules was validated through whole-cell proteomics in patient-derived neurons, revealing their disruption in 22q11.2 deletion carriers diagnosed with SCZ. We demonstrated their therapeutic potential by showing that these modules are targets of GSK3 inhibition using phosphoproteomics. Our findings present a foundational model that integrates genetic, transcriptional, and proteomic perturbations in SCZ. This model provides a cohesive framework for understanding how polygenic and multimodal perturbations affect neuronal pathways in the human brain, as well as a data-driven pathway resource for identifying potential drug targets to reverse disruptions observed in these neuronal networks.

‘Protein interaction networks in the vasculature prioritize genes and pathways underlying coronary artery disease‘ has been published in Communications Biology.

‘Using brain cell-type-specific protein interactomes to interpret neurodevelopmental genetic signals in schizophrenia‘ has been published in iScience.

In this study, we performed interaction proteomics for 6 schizophrenia risk genes prioritized from GWAS loci in human induced excitatory neurons. The resulting protein-protein interaction network is enriched for common variant risk of schizophrenia across both European and East Asian ancestries and can complement fine-mapping and eQTL data to prioritize genes in GWAS loci. We observed convergent genetic signals in the HCN1 sub-network, which is enriched for common variant risk of schizophrenia and contains proteins (HCN4 and AKAP11) enriched for rare protein-truncating mutations in individuals with schizophrenia and bipolar disorder. Our findings in this study (and the sister study on autism spectrum disorders) showcase brain cell-type-specific protein interactomes as an organizing framework to facilitate interpretation of genetic and transcriptomic data in neuropsychiatric disorders.

‘Protein interaction studies in human induced neurons indicate convergent biology underlying autism spectrum disorders‘ has been published in Cell Genomics. Read more about our findings in Greta’s tweetorial.

At this year’s World Congress of Psychiatric Genetics, Greta presented a talk titled “Combining Proteomics and Genetics to Elucidate the Molecular Mechanisms Underlying Neurodevelopmental and Neuropsychiatric Diseases in Human Neurons”. In this presentation, Greta highlighted the BINe projects in which we combined neuroscience, genetics, and proteomics approaches to understand schizophrenia and ASD.

The BINe team posted two preprints on mexRxiv showcasing human brain cell-type-specific protein interaction data as a rich resource for understanding the genetics and neurobiology of schizophrenia and autism spectrum disorders (ASD).

The Novo Nordisk Foundation and Broad Institute of MIT and Harvard are launching a new research Center for Genomic Mechanisms of Disease, an initiative that will accelerate efforts to mine genetic data for insights into disease mechanisms – and eventually rationally designed treatments.

Supported by a $47.5 million commitment from the Novo Nordisk Foundation, the Center will facilitate close collaborations between the Broad Institute and Danish researchers investigating the genetics and gene regulation of common complex disease, with an initial focus on type 2 diabetes and obesity. The Center will align with existing international efforts, data sharing, methodology, and tools to contribute to the roadmap of the International Common Disease Alliance, and working in common cause with investigators from other large-scale efforts such as the Accelerating Medicines Partnership in Common Metabolic Diseases and the Impact of Genomic Variation on Function consortia.

The research collaboration of the Center aims to advance patient-centered research and precision medicine. The Center will establish an exchange program to provide opportunities for Danish scientists to study genomic technologies at the Broad Institute. In turn, these collaborations will catalyze and contribute to expanding biomedical research in Denmark. The new Center will be directed by Kasper, who has played an important role in the planning and development of the initiative since it was conceived in 2019. His whole team is particularly excited and proud to be part of this exceptional research opportunity!

We are hiring for several positions. Email lage.kasper@mgh.harvard.edu with a CV and cover letter if you are interested in hearing about our openings.

‘Translating polygenic risk scores for clinical use by estimating the confidence bounds of risk prediction’ has been published in Nature Communications. This paper describes a machine learning method (MCCP) that can be used to estimate personalized confidence levels of disease risk prediction based on polygenic risk scores, thereby enhancing their utility and interpretability in clinical settings.