Visualizing novel connections and genetic similarities across diseases using a network-medicine based approach

Abstract

Understanding the genetic relationships between human disorders could lead to better treatment and prevention strategies, especially for individuals with multiple comorbidities. A common resource for studying genetic-disease relationships is the GWAS Catalog, a large and well curated repository of SNP-trait associations from various studies and populations. Some of these populations are contained within mega-biobanks such as the Million Veteran Program (MVP), which has enabled the genetic classification of several diseases in a large well-characterized and heterogeneous population. Here we aim to provide a network of the genetic relationships among diseases and to demonstrate the utility of quantifying the extent to which a given resource such as MVP has contributed to the discovery of such relations. We use a network-based approach to evaluate shared variants among thousands of traits in the GWAS Catalog repository. Our results indicate many more novel disease relationships that did not exist in early studies and demonstrate that the network can reveal clusters of diseases mechanistically related. Finally, we show novel disease connections that emerge when MVP data is included, highlighting methodology that can be used to indicate the contributions of a given biobank.

Figure 1

Phenotypic network assembled from GWAS catalog. Network in which nodes are traits that are connected with others to which they share genetic variants in the GWAS Catalog. The network communities detected are highlighted and labeled (A)–(H). Node colors represent disease categories and node size reflects connectivity in the network. The top high degree nodes are labeled 1–10 and their respective names are shown in Table 1. Only significant edges are shown (FDR < 0.05), the edge width indicates the overlap of variants between a pair of phenotypes (Jaccard Index), and lighter shade edges connect nodes in different communities.

Figure 2

Degree distribution of phenotypic network. Log-binned degree distribution of the phenotypic network using a log–log scale. A power-law distribution which is a feature of scale-free networks. K represents the average degree of the bin where bin has size 2^n-1. pK is obtained from the number of nodes found in the bin divided by the width of the bin.

Figure 3

Degree distribution by trait category. Trait categories are defined by the EFO ontology system parent terms.

Figure 4

Network community ‘E’ characterized by immune-related disorders. Focused subgraph of community E from the Phenotypic Network. The most connected diseases in the community are Crohn’s disease (1), rheumatoid arthritis (2), ulcerative colitis (3), psoriasis (4), and lupus (5).

Figure 5

Network community ‘A’ characterized by vascular disorders. After removal of traits unrelated to diseases from the visualization, the most connected nodes in the community were coronary heart disease (1), stroke (2), coronary artery disease (3), metabolic syndrome (4), cardiovascular disease (5), hypertriglyceridemia (6), gout (7), chronic kidney disease (8), diabetes mellitus (9), and atrial fibrillation (10).

Figure 6

Disease connections that emerge from MVP data. Subgraph containing 196 traits and 297 edges that were formed only by the inclusion of genetic variant associations from the Million Veteran Program.

Figure 7

Community correlation between notes of different ancestry networks. Histogram of Pearson product-moment correlation coefficients for shared community members between the same trait in different ancestry networks. The count represents number of traits that have the given correlation coefficient.