【AI Seminar】2024.09.24 Decoding Biological Language in Membrane Proteins: Combining Protein Language Model Embeddings with Multi-Window Scanning Deep Learning for Functional Identification - Yu-Yen Ou Prof

Decoding Biological Language in Membrane Proteins: Combining Protein Language Model Embeddings with Multi-Window Scanning Deep Learning for Functional Identification



Topic: Decoding Biological Language in Membrane Proteins:  Combining Protein Language Model Embeddings with Multi-Window Scanning Deep Learning for Functional Identification

Speaker: Yu-Yen Ou Prof. of Dept. of Computer Science and Engineering,Yzu University

Time: 2024/09/24(Tue) 14:00-16:00

Venue:  The Management Building, 11F,

Chang Gung University, AI Lecture Hall

Join Online: https://gqr.sh/awHN

 

About the Speaker:

Yu-Yen Ou earned his Ph.D. in Computer Science from National Taiwan University in 2005 and is currently a professor in the Department of Computer Science and Engineering at Yuan Ze University. His research interests include machine learning and bioinformatics, particularly focusing on protein sequence function prediction and identification. Since 2016, he has pioneered the integration of deep learning and natural language processing in membrane protein sequence analysis.

Professor Ou has received the 21st Outstanding Professor Award from the Far Eastern Y. Z. Hsu Science and Technology Memorial Foundation and several research grants from Taiwan's Ministry of Science and Technology.

 

Abstract:

This presentation will explore the integration of protein language pre-training models with multi-window scanning deep learning techniques to decode and analyze the biological language embedded within membrane protein sequences. We will first utilize protein language models such as ProtTrans or ESM-2 to transform protein sequences into high-dimensional vector embeddings, capturing the intricate biological language within these sequences. Following this, we will employ multi-window convolutional neural networks (MCNN) to extract features across various scales, enabling the identification of membrane protein functions based on these language features. This innovative approach, combining language models with multi-scale analysis, not only enhances our understanding of membrane and transporter proteins but also offers new perspectives and potential applications in the field of bioinformatics.

 

 

Organizers: College of Intelligent Computing & Artificial Intelligence Research Center


No registration needed.