International Journal of Molecular Medical Science, 2024, Vol.14, No.5, 274-292 http://medscipublisher.com/index.php/ijmms 280 Kundaje et al., 2015). Pan-cancer studies and integrative analyses were also performed (Hoadley et al., 2014; Hoadley et al., 2018). These studies demonstrate that multi-scale or multi-platform genomic research is superior to single-scale research in cancer studies. In research of Tong et al., an integrative prognostic model for colon cancer was developed by combining clinical and multi-omics data, resulting in improved prognostic performance (Tong et al., 2020). Additionally, Li and Sun (2022) proposed a novel method for integrating gene expression, copy number variation, DNA methylation, and miRNA data to identify cancer biomarkers, demonstrating the effectiveness of data fusion in multi-omics studies. 4.3 Tools and software for multi-omics data integration Several tools and software have been developed to facilitate the integration and analysis of multi-omics data. These tools often provide user-friendly interfaces and robust algorithms to handle the complexity of multi-omics datasets. For example, Subramanian et al. reviewed various tools and methods for multi-omics data integration, highlighting their applications in disease subtyping, biomarker prediction, and data interpretation (Table 1). Additionally, Yang et al. introduced the MDICC model, which integrates new affinity matrix and network fusion methods for clustering and identifying cancer subtypes, demonstrating the effectiveness of specialized software in multi-omics data integration. In summary, the integration of multi-omics data is a powerful approach to uncovering the complex mechanisms underlying diseases such as colon cancer. By employing various statistical and computational methods, including correlation-based approaches, machine learning techniques, network-based integration, and data fusion, researchers can enhance the identification of early biomarkers and improve disease prognosis. The development and utilization of specialized tools and software further facilitate these integrative analyses, paving the way for advancements in precision medicine. 5 Identification of Early Colon Cancer Biomarkers 5.1 Genomic alterations as biomarkers Genomic alterations, including mutations and somatic copy number variations, play a crucial role in the identification of early colon cancer biomarkers. For instance, the integration of multi-omics data has revealed specific gene mutations such as TERT and ERBB4, which are associated with improved survival in immunotherapy-treated colon cancers (Elsayed et al., 2022). Ge et al. (2019) identified a four-gene signature (ACVR2A, APC, DOCK2, and POLE) as a strong predictor of survival in high-mutant colorectal cancer (CRC) and found it to be particularly effective in stage II and III colon cancer and MSI-H CRC. Huang et al. (2019) discovered five prognostic genes (MMP1, ACSL6, SMPD1, PPARGC1A, and HEPACAM2), which could provide valuable insights for further research and clinical treatment. Additionally, genes like SLK, which exhibit high missense mutation rates, have been identified as potential prognostic biomarkers across various cancers, including colon cancer (Zhao et al., 2020). In addition, the CNV profiles of 159 genes could be used to predict prognosis of colon cancer patients (Yang et al., 2020). 5.2 Transcriptomic signatures Transcriptomic analysis at the single-cell level has identified differentially expressed genes that serve as marker genes for various colon cancer subtypes. These marker genes show significant specificity compared to normal colon cells, particularly those that are upregulated in tumors (Sun et al., 2021). Deng identified a novel long non-coding RNA (PiHL, p53 Inhibitory LncRNA) as a negative regulator of p53, which is significantly upregulated in colorectal cancer (CRC) and serves as an independent predictor of poor prognosis in CRC (Deng et al., 2020). Moreover, transcriptomic profiles, including mRNA and miRNA expression, have demonstrated high prognostic performance in colon cancer, often outperforming other omics profiles (Zhu et al., 2017). Jacob et al. (2017) used miRNA expression profiles as biomarkers for prognosis in stage II and III colon cancer and identified a 16-miRNA signature as a reliable prognostic marker for stratifying stage II and III colon cancer patients into low and high recurrence risk groups. We also identified 268 genes that are highly expressed in colon cancer tissue, and 13 genes that are associated with colon cancer patient prognosis (Figure 2).
RkJQdWJsaXNoZXIy MjQ4ODYzNQ==