Multi-omics approaches in cancer research with applications in tumor subtyping, prognosis, and diagnosis

While cost-effective high-throughput technologies provide an increasing amount of data, the analyses of single layers of data seldom provide causal relations. Multi-omics data integration strategies across different cellular function levels, including genomes, epigenomes, transcriptomes, proteomes, metabolomes, and microbiomes offer unparalleled opportunities to understand the underlying biology of complex diseases, such as cancer. We review some of the most frequently used data integration methods and outline research areas where multi-omics significantly benefit our understanding of the process and outcome of the malignant transformation. We discuss algorithmic frameworks developed to reveal cancer subtypes, disease mechanisms, and methods for identifying driver genomic alterations and consider the significance of multi-omics in tumor classifications, diagnostics, and prognostications. We provide a comprehensive summary of each omics strategy's most recent advances within the clinical context and discuss the main challenges facing their clinical implementations. Despite its unparalleled advantages, multi-omics data integration is slow to enter everyday clinics. One major obstacle is the uneven maturity of different omics approaches and the growing gap between generating large volumes of data compared to data processing capacity. Progressive initiatives to enforce the standardization of sample processing and analytical pipelines, multidisciplinary training of experts for data analysis and interpretation are vital to facilitate the translatability of theoretical findings.