Dissecting big RNA-Seq cancer data using machine learning to find disease-associated genes and the causal mechanism

The exploitation of machine learning (ML) methods to analyze RNA-Seq data has opened new avenues of unprecedented power to investigate biological systems. Although RNA-Seq is only a decade old, the application of ML to analyze such data has already overshadowed the traditional microarray approach, especially in diseases like cancer, where such applications have taken exponential growth. For a comprehensive understanding of these applications, in this book chapter, we have discussed the basic principles of ML techniques. This understanding is then further enacted on breast cancer data to achieve biological insights, where we built a hybrid model consisting of supervised and unsupervised techniques. With this model’s help, we identified the deeply associated genes in the disease and hypothesized that they have a role in the causal mechanism of the disease. Further, the protein–protein interaction network analysis of the DEGs reveals that such proteins are highly significant as they are amidst the topologically strong proteins in the breast cancer network.