A survey on Multi-Omics Analysis for Breast Cancer Classification and Precision Medicine

Breast cancer has surpassed lung cancer as the most frequently diagnosed cancer, with an estimated 2.3 million new cases out of $\mathbf{1 9. 3}$ million cases in 2020. Breast cancer itself has heterogeneous characteristics both in terms of genotype and phenotype. This paper provides a survey on multi-omics analysis for breast cancer. Multi-omics approaches can provide a holistic understanding of disease mechanisms, identify biomarkers, and find effective therapies. The analysis results from multi-omics and single-omics approaches could be similar or different biomarkers, depending on the biological complexity of the condition under study, such as breast cancer, and the data integration capabilities of different omics platforms. Feature selection or dimensionality reduction aims to reduce the number of multi-omics variables. A brief concept of data integration is achieved through deep learning. First, preprocess and clean the multi-omics data, then use conventional feature selection techniques or feature reduction methods. The selected multiomics variables were combined into one data set for data integration. Finally, the integrated data is analyzed using classification, regression, and clustering.