![]() ![]() ![]() However, current graphlet-based methods ignore neighbourhood infor- mation (i.e., what nodes are connected). The state-of-the-art topological node and network descriptors are based on graphlets, induced connected subgraphs of different shapes (e.g., paths, triangles). Topo- logical analysis of these networks (i.e., the analysis of their structure) has led to break- throughs in biology and medicine. Recent biotechnological advances have led to a wealth of biological network data. Finally, we used SBM-GNN to perform a pan-cancer analysis, where we found genes and pathways directly involved with the hallmarks of cancer controlling genome stability, apoptosis, immune response, and metabolism. Experimental analysis of synthetic datasets showed that our model can correctly predict genes associated with cancer and recover relevant pathways, while outperforming other state-of-the-art methods. Here we take advantage of recent advances in graph neural networks, combined with well established statistical models of network structure, to build a new model, called Stochastic Block Model Graph Neural Network (SBM-GNN), which predicts cancer driver genes and cancer mediating pathways directly from high-throughput omic experiments. Since most biological processes are the results of the interaction of multiple genes, it is then conceivable that tumorigenesis is likely the result of the action of networks of cancer driver and non-driver genes. The advent of high-throughput omic technologies has enabled the discovery of a significant number of cancer driver genes, but recent genomic studies have shown these to be only necessary but not sufficient to trigger tumorigenesis. The identification of genes and pathways responsible for the transformation of normal cellsinto malignant ones represents a pivotal step to understand the aetiology of cancer, to characterise progression and relapse, and to ultimately design targeted therapies. Supplementary material are available at Bioinformatics online. Hierarchical HotNet is a robust algorithm for identifying altered subnetworks across different ‘omics datasets. On somatic mutation data from The Cancer Genome Atlas, Hierarchical HotNet outperforms other methods and identifies significantly mutated subnetworks containing both well-known cancer genes and candidate cancer genes that are rarely mutated in the cohort. We evaluate the performance of Hierarchical HotNet and several other algorithms that identify altered subnetworks on the problem of predicting cancer genes and significantly mutated subnetworks. Hierarchical HotNet assesses the statistical significance of the resulting subnetworks over a range of biological scales and explicitly controls for ascertainment bias in the network. We introduce Hierarchical HotNet, an algorithm that finds a hierarchy of altered subnetworks. In these and other applications, the underlying computational problem is to identify altered subnetworks containing genes that are both highly altered in an ‘omics dataset and are topologically close (e.g. For example, protein–protein interaction networks have been used to analyze gene expression data, to prioritize germline variants, and to identify somatic driver mutations in cancer. The analysis of high-dimensional ‘omics data is often informed by the use of biological interaction networks. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |