Yun-Tao Liu

This paper introduces spIsoNet, a deep-learning method to correct preferred-orientation bias in cryoEM reconstructions, demonstrating its effectiveness on influenza HA trimer datasets.

IsoNet2 reconstructs high-resolution 3D cellular structures from cryogenic electron tomography without averaging.

This study identifies a new D-capsid type in herpesvirus assembly, revealing structural dynamics of DNA translocation and clarifying the origins of empty and DNA-filled capsids.

This paper introduces spIsoNet, a deep-learning tool that improves 3D reconstruction isotropy and particle alignment in cryoEM by addressing the preferred orientation problem without altering specimen preparation.

This study discovers that native MCC from Leishmania tarentolae forms filaments, determines its structure, and proposes a new model for its catalytic mechanism.

DeepMAPS infers biological networks from single-cell multi-omics data using a heterogeneous graph transformer, outperforming existing tools in cell clustering and pathway prediction.

The paper identifies the in situ structure of GABAAR in hippocampal inhibitory synapses and describes its hierarchical organization, revealing mesophasic self-organization that balances stability and flexibility.