MicroRNAs in oncology: a translational perspective in the era of AI

Comprehensive Summary

microRNA (miRNA) has been utilized recently for a more comprehensive understanding of normal physiology and most diseases, especially tumorigenesis. Jurj et al. analyzed the use of machine learning techniques to utilize miRNA as tumor biomarkers for identification and clinical trials. Studies utilizing miRNA biomarkers for diagnostics specifically investigate miRNA transport mechanisms such as transcription factor networks, epigenetic modifications, RNA editing, and core miRNA biogenesis machinery. miRNA was found to act as crucial regulators as both oncogenic factors and tumor suppressors, sometimes even with dual roles. AI was utilized to identify multi-miRNA signatures for cancer subtypes and early detection via creation of miRNA indexes and signatures. These miRNAs are encapsulated in extracellular vesicles which can also be studies to look at diagnostics and cancer proliferation. Specific carcinomas have been accurately diagnosed in studies when using panels of miRNA derived from specific bodily fluids and carriers. Machine learning has been able to integrate huge, multi-modal datasets comprised of miRNA, genetics, and clinical findings to uncover patterns for diagnostic models and subtyping. This allows for future therapeutics development with possible improved delivery systems of treatments with miRNA.

Outcomes and Implications

This review allows for the view of cancer as a systems-level dysregulation through the mass targeting of miRNAs, allowing for focus on gene regulatory networks and pathway crosstalk. In addition, these dynamic signatures reflecting tumor origin, aggressiveness, possible treatment response, and early disease detection can be further studies for future usage in non-invasive oncology. These machine learning models can be use for predictive measures in prognoses and proper patient stratification for optimal treatment plans. these miRNAs can act as master regulators with a broad range of pathways; this has a possibility to be used for therapeutics once becoming more personalized alongside approved clinical treatments.