Zhengming Ding

This paper introduces a nonlinear mixed model (NMM) for fMRI analysis to better capture nonlinear relationships and individual variability, showing improved model fit compared to linear models.

The paper develops M2AE, a multi-task multi-view model to impute blood SCFA levels using gut metagenomic data, dietary features, and host characteristics, showing improved performance over existing methods.

This paper introduces DemoVAE, a variational autoencoder that generates synthetic fMRI data conditioned on demographics and removes demographic confounds from fMRI features.

The paper introduces PTM-TBA, a software pipeline that identifies and localizes post-translational modifications in proteoforms using top-down and bottom-up mass spectrometry combined with UniProt annotations.

MORONET is a multi-omics integration method using graph convolutional networks for biomedical classification, outperforming existing approaches and identifying disease-related biomarkers.