Mol Genet Metab. 2017 May;121(1):43-50. doi: 10.1016/j.ymgme.2017.03.004.

Type-2 diabetes-associated variants with cross-trait relevance: Post-GWAs strategies for biological function interpretation.

PubMed link:



To use genome-wide association studies (GWAs) results towards therapeutic targets discovery there is a need for high-throughput methods to interpret GWAs summary statistics, integrate data from multiple omics analyses and facilitate follow-up studies. We developed a pipeline for interpretation of GWAs results in a biological context that is useful for a Type-2 diabetes (T2D) drug development perspective. Using 13 GWAs datasets from T2D, glycemic traits, obesity, lipids and cardiovascular disease traits, we detected T2D associated genes with genetic variants that have an effect across these traits (Figure 1). The characterization of the wider phenotypic impact of these variants provides clues to the consequences of therapeutic perturbations of the cognate pathway.



Figure 1. Systems Genomics: we searched for genetic variants associated with T2D and either of other 12 related traits (fasting glucose, fasting insulin, 2h glucose, HbA1c, obesity class I (BMI ≥ 30-34.9), obesity class II (BMI ≥ 35-39.9), obesity class III (BMI ≥ 40 kg/m2), coronary artery disease, TC, HDL, LDL, and TG.  Summary statistics from GWAs analyses included in this study were obtained as open source data from the following consortia: DIAbetes Genetics Replication and Meta-analysis (DIAGRAM), the Meta-Analyses of Glucose and Insulin-related traits Consortium (MAGIC), The Genetic Investigation of ANthropometric Traits (GIANT), Coronary ARtery DIsease Genome wide Replication and Meta-analysis (CARDIOGRAM), and Global Lipids Genetics Consortium (GLGC).


For biological interpretation (Figure 2), we annotated these genes for tissue specificity, gene expression changes in islets of T2D patients versus healthy subjects, and built networks to understand their connectivity and common pathways. We also annotated the functional relevance of the identified variants based on their known impact on gene expression or protein function. Using the above information we ranked the genes based on their relevance to the disease process. This can contribute to: (i) understand how to modulate a gene of interest to achieve a desired effect on a biological pathway, (ii) design follow up studies to determine whether a gene perturbation is a cause or consequence of a human disease process, and (iii) understand its mechanism of action.



Figure 2. Clinical translation process. Workflow from identification of relevant genetic variants to functional annotation and target translational strategy.


In result, we identified that many of the T2D1 genes with cross traits effects, (i) were encoding for closely interacting proteins, (ii) had functional studies supporting their relevance for the associated traits, (iii) were cis eQTLs associated with transcript levels in different tissues, and (iv) mapped to genes that were differentially expressed in the islets of T2D patients versus healthy individuals.

In summary, we demonstrated how systems genomics and network medicine approaches can shed light into T2D GWAs discoveries, translating findings into a more therapeutically relevant context.

Open-source bioinformatics tools’ URLs: DEPICT,; GTEx portal,; HaploReg V3 tool,; HumanMine,; Locus Zoom,




  1. Morris et al. Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. Nat Genet. (2012) 44(9):981-90. DOI 10.1038/ng.2383