Fei Sun

This study reveals the subnanometer architecture of basal bodies in mammalian cells, identifies new components, and proposes a mechanism for central pair microtubule nucleation during ciliogenesis.

cryoAgent improves cryo-EM image processing by enabling autonomous, adaptive workflows that enhance reconstruction quality and identify new structural states.

The paper introduces PLCT, a method for depth-resolved structural analysis of multilayered tissues using cryo-ET, and demonstrates its ability to resolve subcellular structures in retinal tissue.

CryoForge is an evidence-gated agent that improves cryo-EM model building by selectively accepting or rejecting edits based on structural evidence, achieving validated improvements in most test cases.

The paper identifies TRiC/CCT as a chaperonin in mammalian sperm flagella that folds locally translated proteins, supporting flagellar function and fertility.

AI-HEAD automatically and rapidly determines helical assembly structures by indexing diffraction data and calculating parameters with semi-automated sorting.

This study shows PCL proteins phase-separate to form PRC2 condensates, regulated by CpG islands disrupting their intramolecular interactions.

This study shows that CFAP77 is essential for connecting axonemal A and B tubules in mammalian sperm, as its loss disrupts the CFAP77-CCDC105-TEX43 complex and impairs sperm motility.

This study investigates whether pulsed-electron imaging in ultrafast cryo-EM mitigates radiation damage, finding no significant mitigation effect despite varying experimental conditions.

This study identifies ANKRD5 as a critical component of the N-DRC in sperm axonemes, showing its role in maintaining structural integrity and motility, with implications for male fertility.

DomainSeeker identifies protein domains in cryo-ET maps using AlphaFold2 structures and Bayesian inference, improving accuracy and resolution robustness.

The paper solves the in-cell structure of mouse sperm central apparatus at sub-nanometer resolution and identifies proteins critical for its asymmetric assembly.

DeepETPicker improves 3D particle picking in cryo-electron tomography with weakly supervised deep learning, achieving higher accuracy and speed than existing methods.

The paper determines the 8 Å cryo-EM structure of the Xenopus laevis nuclear pore complex nuclear ring using AlphaFold2 to build a pseudoatomic model.

The paper determines the architecture of the nuclear ring in Xenopus laevis NPC using cryo-EM, homology modeling, and cell experiments, identifying key proteins and their roles in NPC assembly.

The paper solves the 8 Å cryo-EM structure of the Xenopus laevis nuclear pore complex cytoplasmic ring using AlphaFold2, revealing detailed structural features of key components.

The paper introduces the SmartBac system for efficiently producing large protein complexes in insect cells, demonstrating its application in expressing six multiprotein complexes.