Nucleic acid variant analysis is a critical tool for deciphering gene functions, studying disease mechanisms, and enabling precision medicine, playing an irreplaceable role in early tumor diagnosis, postoperative monitoring, and personalized medication guidance. However, current methods face bottlenecks such as low throughput, insufficient sensitivity, and difficulty in multiplex co-detection, particularly for trace, low-abundance nucleic acid variants in liquid biopsy. This study focuses on the development and application of high-throughput, intelligent, ultra-parallel nucleic acid analysis technology. By proposing a "ribozyme-thermodynamic" nucleic acid molecular probe hybridization amplification model and incorporating a thermodynamic base-stacking intelligent computation method, we constructed a new model for accurately predicting nucleic acid hybridization and amplification. Based on this, we intelligently designed high-throughput nucleic acid probes, breaking the throughput limitations of conventional methods and achieving 100,000-plex ultra-parallel amplification in a single tube, increasing throughput by 1,000 times. We also established a highly sensitive detection system for various mutation types, including low-frequency point mutations, fusion genes, deletions, and insertions, and identified 42 novel fusion gene markers for cervical cancer, preliminarily validating their tumor-promoting mechanisms and drug sensitivity. Furthermore, we developed a high-throughput method for detecting trace low-frequency mutations, capable of detecting 87 mutations using only a single probe, reducing the detection limit by 10-fold, shortening assay time by 4-fold, and cutting costs by 50-fold. This method has been successfully applied to the precise subtyping and personalized medication guidance of colorectal cancer, liver cancer, and other tumors, and the related technology has been translated into clinical molecular diagnostics, providing important tools and methods for precision oncology with significant clinical translation potential.

Speaker

Prof. Ping Song

School of Biomedical Engineering, SJTU

Time

 2026.5.22 12:00-13:30