Nature Sub-Journal: Researchers use artificial intelligence to analyze glucose metabolism data

Researchers at the University of California, San Diego have developed a tool that can analyze sugar comics datasets using interpretable artificial intelligence (AI) systems and other machine learning methods

To introduce GlyCompare, the team demonstrated their ability to enhance the comparison of glycomics data sets by revealing hidden relationships between glycans in a variety of situations, including gastric cancer tissue

"We applied GlyCompare to cancer tissues, and the results showed that although cancer-specific glycans could not be found using standard statistical methods, new biomarkers appeared after processing with our method," University of California, San Diego, Bioengineering and Pediatrics professor Nathan Lewis said he is the corresponding author of the paper

In another analysis, the team showed that the method greatly improves statistical power, for example, half of the sample is required to obtain the same power to detect biomarkers

One of the key points of the GlyCompare method is that it focuses on the biological steps required to synthesize the subunits of the polysaccharide, not just the entire polysaccharide itself, which greatly improves the accuracy of the statistical analysis of sugar data

Bokan Bao and Benjamin P.

Paper title

Correcting for sparsity and interdependence in glycomics by accounting for glycan biosynthesis