DGH-GO: Dissecting the Genetic Heterogeneity of complex diseases using Gene Ontology

Complex diseases such as neurodevelopmental disorders (NDDs) lack biological markers for their diagnosis and are phenotypically heterogeneous, which makes them difficult to diagnose at early-age. The genetic heterogeneity corresponds to their clinical phenotype variability and, because of this, complex diseases exhibit multiple etiologies. The multi-etiological aspects of complex-diseases emerge from distinct but functionally similar group of genes. Different diseases sharing genes of such groups show related clinical outcomes that further restrict our understanding of disease mechanisms, thus, limiting the applications of personalized medicine or systems biomedicine approaches to complex genetic disorders. Here, we present an interactive and user-friendly application, DGH-GO that allows biologists to dissect the genetic heterogeneity of complex diseases by stratifying the putative disease-causing genes into clusters that may lead to or contribute to a specific disease traits development. The application can also be used to study the shared etiology of complex-diseases. DGH-GO creates a semantic similarity matrix of putative disease-causing genes or known-disease genes for multiple disorders using Gene Ontology (GO). The resultant matrix can be visualized in a 2D space using different dimension reduction methods (T-SNE, Principal component analysis and Principal coordinate analysis). Functional similarities assessed through GO and semantic similarity measure can be used to identify clusters of functionally similar genes that may generate a disease specific traits. This can be achieved by employing four different clustering methods (K-means, Hierarchical, Fuzzy and PAM). The user may change the clustering parameters and see their effect on stratification results immediately. DGH-GO was applied to genes disrupted by rare genetic variants in Autism Spectrum Disorder (ASD) patients. The analysis confirmed the multi-etiological nature of ASD by identifying the four clusters that were enriched for distinct biological mechanisms and phenotypic terms. In the second case study, the analysis of genes shared by different NDDs showed that genes involving in multiple disorders tend to aggregate in similar clusters, indicating a possible shared etiology. In summary, functional similarities, dimension reduction and clustering methods, coupled with interactive visualization and control over analysis allows biologists to explore and analyze their datasets without requiring expert knowledge on these methods. The source code of proposed application is available at https://github.com/Muh-Asif/DGH-GO Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=101 SRC="FIGDIR/small/513077v1_ufig1.gif" ALT="Figure 1"> View larger version (27K): org.highwire.dtl.DTLVardef@1a79e61org.highwire.dtl.DTLVardef@18ef52dorg.highwire.dtl.DTLVardef@827df4org.highwire.dtl.DTLVardef@11b4049_HPS_FORMAT_FIGEXP M_FIG C_FIG