Akihiro Kitani

This paper evaluates when site-specific glycoform compositions can resolve biosynthetic rates, showing that sufficient panel size, separable responses, and model fit are required for accurate inference.

GlycoMeSH links glycan structures to biomedical contexts through a database and enrichment workflow, improving association recovery and coverage over existing methods.

GlycanGT is a foundation model for glycans that uses a graph transformer to improve classification and prediction of glycan structures and properties.

This study uses deep learning to integrate proteomics and PPI data, identifying astrocytes and GPNMB+ microglia as novel moderators of Aβ-tau interaction in early Alzheimer's disease.

The paper uses machine learning on multi-cohort transcriptomic and proteomic data to identify molecular subtypes of cognitive resilience in Alzheimer's disease and finds two major subgroups with distinct proteomic and neuroimaging features.