Machine Learning-based Multi-Omics Data Analysis for Cancer Diagnosis and Prediction

Recent progress in bioinformatics has made it possible to extract useful information from massive datasets of gene expression. Researchers are increasingly concerned with how to identify important genes among these massive databases. MicroRNAs were identified as short non-coding RNAs that regulate gene expression by binding to the 3' UTR of target messenger RNAs and so inhibiting protein synthesis in accordance with the findings of the in-depth study. A variety of diseases, including cancer, have been linked to deregulation of microRNAs since their discovery. Using multi-omics data analysis, this study intends to delve into the characteristics and roles of these microRNAs and genes. MicroRNAs and genes are singled out for their functional and biological relevance across a wide range of cancer types using state-ofthe- art machine learning algorithms. Furthermore, ligand binding activity prediction of Cyclin- Dependent Kinases is predicted using machine learning approaches. Eukaryotic cell division is coordinated by a group of proteins called cyclin-dependent kinases. Cyclin-dependent kinase crystal structures often reveal potential molecular processes of ligand binding.