Seminars
May 29, 2026
Building fully automated cryo-EM data processing pipeline with foundation models
Speaker: Dr. Huaizong Shen
Principal Investigator & Deputy Director, Institute of Bio-Architecture and Bio-Interaction, Shenzhen Medical Academy of Research and Translation (SMART)
School of Biomedical Sciences cordially invites you to join the following seminar:
Date: 29 May 2026 (Friday)
Time: 3:00 pm – 4:00 pm
Venue: Lecture Theatre 1, G/F, William M.W. Mong Block, 21 Sassoon Road
Host: Professor Keda Zhou
Biography
Dr. Huaizong Shen leads the Laboratory of Structual Biology and Artificial Intelligence at Institute of Bio-Architecture and Bio-Interaction, Shenzhen Medical Academy of Research and Translation (SMART). His research sits at the intersection of structural biology and artificial intelligence, focusing on deciphering the structure and function of pharmaceutically important ion channels and membrane receptors. Meantime, by developing and applying novel AI-driven methods, his team aims to democratize the frontier cryo-EM technology and accelerate the discovery of novel therapeutic targets and drug precursors.
Dr. Shen has authored landmark studies on voltage-gated sodium and potassium channels, as long as AI-driven cryo-EM democratization in top-tier journals. He is a recognized young leader, honored with China's Excellent Young Scientist Fund and National High-Level Talent Special Support Program.
Abstract
Cryo-electron microscopy (Cryo-EM) has emerged as a cutting-edge technology for determining high-resolution structures of biological macromolecules. However, its widespread application is limited by the stringent requirement for specialized expertise in cryo-EM data processing. Leveraging an unsupervised learning strategy, we have developed a foundation model for Cryo-EM image evaluation, termed Cryo-IEF, by pre-training on a large-scale dataset of approximately 134 million particle images. This model demonstrates outstanding performance on various Cryo-EM data processing tasks, including classification of particles with different structures, clustering of particle orientations, and assessment of image quality. Based on the foundation model, we constructed a fully automated single-particle Cryo-EM data processing pipeline, CryoWizard, which achieves a proficiency that rivals human experts. The synergistic innovation of the Cryo-IEF model and the CryoWizard pipeline represents a significant step towards making Cryo-EM technology more accessible, efficient, and robust, thereby promising to advance research across diverse fields in the life sciences.
All are welcome.
New Releases
Jun 8, 2026
HKUMed School of Biomedical Sciences and Votee AI sign MoU to launch student internships and joint R&D in applied AI
May 28, 2026
HKUMed develops innovative tool to repair genetic errors, offering new hope for neurodegenerative diseases
Mar 9, 2026