Projects under the Novo Nordisk Foundation Center for Genomic Mechanisms of Disease

The overarching goal of the Novo Nordisk Foundation Center is to mine genetic data for insights into disease mechanisms that will lead to the development of new treatments. In a collaboration between the Broad Institute and Danish institutions, the Center will primarily focus on two transformative research directions:

  1. Genetic Basis of Human Disease: investigate the genetic and cellular basis of type 2 diabetes and obesity. We will use our deep knowledge gained from the study of diverse polygenic disorders to identify and characterize the key cell types and tissues involved in these diseases. To do so, we will apply our framework that consists in computationally prioritizing disease risk genes from genetic data. We will then generate cell-type-specific genomic, epigenomic, transcriptomic, and proteomic data in the relevant cell types and integrate the results to build a comprehensive disease mechanism map for both disorders.
  2. Human Gene Regulatory Map: build gene regulation maps in cell types related to metabolic diseases, such as type 2 diabetes and obesity. We will use our expertise in transcriptional regulation to contribute to the identification of regulatory regions that are perturbed in disease. This will help understand which variants are responsible for transcriptional changes in the tissues and cells that are involved in disease.

Advances made in both directions will allow identifying disease genes and regulatory pathways that could be targeted by new classes of therapeutics in cells and tissues.

The Brain Interaction Network (BINe) project at the Stanley Center for Psychiatric Research

Advances in human genetics have nominated many genes and specific cell types implicated in psychiatric illnesses, but linking these findings with biochemical processes remains challenging. To overcome this bottleneck, we developed an experimental and computational framework to generate the protein-protein interaction networks of schizophrenia, autism, and post traumatic stress disorder (PTSD) risk genes in human excitatory neurons induced from pluripotent stem cells. We observed that the resulting protein networks are often neuron-specific, enriched for genetic risk of the disease, and include many novel interactions not reported in the literature. These results demonstrate the potential of using cell-type-specific protein interaction networks to interpret genetic association studies and prioritize pathways for follow-up functional investigation. Therefore, our data are a rich resource for other scientists in the field to use in the interpretation of their own datasets. We are currently expanding our approach to study additional disease risk genes in multiple brain cell types implicated in psychiatric disorders, while continuing to identify new disease-specific pathways and to validate interactions discovered in our networks.

Figure 1. Example protein interaction network of six index proteins (i.e., proteins used as baits in immunoprecipitation-mass spectrometry experiments). Index proteins and interaction partners are indicated as red and purple nodes, respectively. Size and color intensity of the interaction partner nodes scale with the number of index proteins linked to each partner (distribution shown in upper right pie chart). Edges represent observed interactions that are either known in the InWeb database (blue) or are potentially novel (grey; distribution shown in lower right pie chart).

Bioinformatics & methods development

We have developed various computational methods and tools to perform proteomic-focused data analyses. For instance, Genoppi is an open-source application for analyzing quantitative proteomic datasets and integrating the results with other genetic and transcriptomic data, thereby facilitating data exploration and hypothesis generation. We are currently working to expand the functionalities of Genoppi, as well as developing computational pipelines to jointly analyze the diverse genomic, epigenomic, transcriptomic, and proteomic data types generated across our projects.

Figure 2. Overview of the Genoppi features.