Integration of resol/block-copolymer carbonization and machine learning: A convenient approach for precise monitoring of glycan-associated disorders

Abstract

Ensuring the timely and precise monitoring of severe liver diseases is crucial for guiding effective therapies and significantly extending overall quality of life. However, this remains a worldwide challenge, given the high incidence rate and the presence of strong confounding clinical symptoms. Herein, we applied a convenient and high-yield method to prepare the magnetic mesoporous carbon (MMC-Fe), guided by a composite of resol and triblock copolymer. With the combination of MMC-Fe, high-throughput mass spectrometry, and a simple machine learning algorithm, we extracted N-glycan profiles from various serum samples, including healthy controls, liver cirrhosis, and liver cancer, and from which we screened specific N-glycans. Specifically, the selected N-glycans demonstrate exceptional performance with area under the curve (AUC) values ranging from 0.948 to 0.993 for the detection of liver diseases, including alpha fetoprotein (AFP)-negative liver cancer. Among them, five N-glycans holds potential in monitoring distinctions between liver cirrhosis and AFP-negative liver cancer (AUC values of 0.827–0.842). This study is expected to promote the glycan-based precise monitoring of diseases, not limited to liver disease.