Dear Wang Lu ,

In computer networking, a heterogeneous network is a network connecting computers and other devices where the operating systems and protocols have significant differences.

Article Heterogeneous Network Edge Prediction: A Data Integration Ap...

Regards,

Shafagat

The first decade of Genome Wide Association Studies (GWAS) has uncovered a wealth ofdisease-associated variants. Two important derivations will be the translation of this infor-mation into a multiscale understanding of pathogenic variants and leveraging existing datato increase the power of existing and future studies through prioritization. We explore edgeprediction on heterogeneous networks—graphs with multiple node and edge types—foraccomplishing both tasks. First we constructed a network with 18 node types—genes, dis-eases, tissues, pathophysiologies, and 14 MSigDB (molecular signatures database) collec-tions—and 19 edge types from high-throughput publicly-available resources. From thisnetwork composed of 40,343 nodes and 1,608,168 edges, we extracted features thatdescribe the topology between specific genes and diseases. Next, we trained a model fromGWAS associations and predicted the probability of association between each protein-cod-ing gene and each of 29 well-studied complex diseases. The model, which achieved 132-fold enrichment in precision at 10% recall, outperformed any individual domain, highlightingthe benefit of integrative approaches. We identified pleiotropy, transcriptional signatures ofperturbations, pathways, and protein interactions as influential mechanisms explainingpathogenesis. Our method successfully predicted the results (with AUROC = 0.79) from awithheld multiple sclerosis (MS) GWAS despite starting with only 13 previously associatedgenes. Finally, we combined our network predictions with statistical evidence of associationto propose four novel MS genes, three of which (JAK2,REL,RUNX3) validated on themasked GWAS. Furthermore, our predictions provide biological support highlightingRELasthe causal gene within its gene-rich locus. Users can browse all predictions online (http://het.io). Heterogeneous network edge prediction effectively prioritized genetic associationsand provides a powerful new approach for data integration across multiple domains. (PDF) Heterogeneous Network Edge Prediction: A Data Integration Approach to Prioritize Disease-Associated Genes. Available from: https://www.researchgate.net/publication/279989646_Heterogeneous_Network_Edge_Prediction_A_Data_Integration_Approach_to_Prioritize_Disease-Associated_Gene

https://discuss.dgl.ai/t/how-to-predict-the-directional-edges-in-a-heterogeneous-network/1885

Article Heterogeneous Network Edge Prediction: A Data Integration Ap...