Project Summary

The methylation patterns of cell-free DNA (cfDNA), released by cells into biofluids, serve as indicators of the epigenetic changes occurring in the entire body. Therefore, sensitive detection of cfDNA plays a crucial role in improving the early detection of cancer. Previous research has demonstrated the successful application of surface-enhanced Raman spectroscopy (SERS) in detecting single DNA base mismatches, specific mutations, and genomic DNA methylation. Building on this knowledge, the objective of this project is to advance the methodology of SERS combined with machine learning (ML) to detect the methylation level of cfDNA.

The project will involve extracting cfDNA from the growth medium of ten cell lines, including seven malignant and three non-tumoral cell lines. The optimization of conditions for effectively adsorbing cfDNA onto silver nanoparticles, which enhances the SERS signal, will be a key focus. Various cations and silver nanoparticles will be tested to achieve this optimization. A partial least square algorithm will be trained and validated to quantify the methylation level of cfDNA using the SERS spectra. Additionally, ML algorithms will be employed to classify cfDNA originating from malignant cells versus non-tumoral cells.

In summary, this proposal aims to leverage the sensitivity, speed, and affordability of SERS as an effective alternative to solid biopsies for comprehensive genomic analysis and early-stage cancer screening.