Genomics and Applied Biology 2024, Vol.15, No.4, 172-181 http://bioscipublisher.com/index.php/gab 174 vector machines are employed to manage the high dimensionality and complexity of omics datasets, particularly in cancer research (Reel et al., 2021; Feldner-Busztin et al., 2023). Additionally, deep learning techniques are increasingly used to predict experimental outcomes and improve the reliability of analytical workflows in proteomics (Mann et al., 2021). The integration of ML in genomic research not only enhances the understanding of biological systems but also supports precision medicine by enabling accurate disease prediction and patient stratification (Mirza et al., 2019; Reel et al., 2021). Figure 1 Role of NGS technology in cancer diagnosis, prognosis, and therapeutics using an integrative omics approach (Adopted from Satam et al., 2023) Image caption: FFPE, formalin-fixed paraffin-embedded; Bx, biopsy; AI, artificial intelligence; Ml, machine learning (Adopted from Satam et al., 2023) 3.2 High-dimensional data analysis The analysis of high-dimensional data is a significant challenge in genomic research due to the large number of features and relatively small sample sizes. Techniques such as dimensionality reduction are crucial for managing this complexity. Methods like principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE), and autoencoders are commonly used to reduce the feature space while preserving essential information (Feldner-Busztin et al., 2023). These techniques help in uncovering genotype-phenotype interactions and identifying true associations while minimizing false positives (Ritchie et al., 2015). The application of these methods is particularly evident in studies utilizing The Cancer Genome Atlas dataset, which provides a rich source of diverse experimental data (Feldner-Busztin et al., 2023). 3.3 Advances in multi-omics data integration Multi-omics data integration is a rapidly evolving field that aims to combine data from various omics layers, such as genomics, transcriptomics, proteomics, and metabolomics, to gain a comprehensive understanding of biological
RkJQdWJsaXNoZXIy MjQ4ODYzMg==