Yijie Zhang

Bonjour Hi! I am a first-year Computer Science Ph.D. student at McGill University and MILA - Québec AI Institute , advised by Prof. Mathieu Blanchette. Prior to this, I obtained my Master's degree from the same department in August 2024.

My research interests lie at the intersection of artificial intelligence and computational biology. Currently, I am focused on developing accurate and lightweight encoders to enhance the understanding of structure-function relationships in various biomolecules.

Aussi, je suis en train d'apprendre le français :)


Education
  • McGill University, MILA
    McGill University, MILA
    School of Computer Science
    Ph.D. Student
    Sep. 2024 - present
  • McGill University
    McGill University
    Master of Science in Computer Science Jan. 2023 - Aug. 2024
News
2025
Our paper on Antibody design was accepted by AAAI 2025 as an Oral Presentation! (Top 4%)
Jan 18
2024
Our paper on RNA design was accepted by Briefings in Bioinformatics!
Dec 02
Happy to get my Master's degree and start my PhD career at McGill and MILA! Merci beaucoup pour tous!
Aug 30
Selected Publications (view all )
dyAb: Flow Matching for Flexible Antibody Design with AlphaFold-driven Pre-binding Antigen
dyAb: Flow Matching for Flexible Antibody Design with AlphaFold-driven Pre-binding Antigen

Yijie Zhang*, Cheng Tan*, Zhangyang Gao*, Yufei Huang, Haitao Lin, Lirong Wu, Fandi Wu, Mathieu Blanchette#, Stan Z Li# (* equal contribution, # corresponding author)

The Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI) 2025 Oral Presentation

We introduce dyAb, a framework leveraging AlphaFold2-driven predictions and combining coarse-grained interface alignment with fine-grained flow matching to simulate dynamic interactions, significantly outperforming existing models and streamlining antibody design.

dyAb: Flow Matching for Flexible Antibody Design with AlphaFold-driven Pre-binding Antigen

Yijie Zhang*, Cheng Tan*, Zhangyang Gao*, Yufei Huang, Haitao Lin, Lirong Wu, Fandi Wu, Mathieu Blanchette#, Stan Z Li# (* equal contribution, # corresponding author)

The Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI) 2025 Oral Presentation

We introduce dyAb, a framework leveraging AlphaFold2-driven predictions and combining coarse-grained interface alignment with fine-grained flow matching to simulate dynamic interactions, significantly outperforming existing models and streamlining antibody design.

R3Design: Deep Tertiary Structure-based RNA Sequence Design and Beyond
R3Design: Deep Tertiary Structure-based RNA Sequence Design and Beyond

Yijie Zhang*, Cheng Tan*, Zhangyang Gao*, Hanqun Cao, Siyuan Li, Siqi Ma, Mathieu Blanchette#, Stan Z Li# (* equal contribution, # corresponding author)

Briefings in Bioinformatics, 2025

R3Design is a tertiary structure-based RNA sequence design method that prioritizes tertiary interactions, significantly outperforming traditional secondary structure-based approaches. By enabling the design, folding, and evaluation of RNA sequences that fold into desired tertiary structures.

R3Design: Deep Tertiary Structure-based RNA Sequence Design and Beyond

Yijie Zhang*, Cheng Tan*, Zhangyang Gao*, Hanqun Cao, Siyuan Li, Siqi Ma, Mathieu Blanchette#, Stan Z Li# (* equal contribution, # corresponding author)

Briefings in Bioinformatics, 2025

R3Design is a tertiary structure-based RNA sequence design method that prioritizes tertiary interactions, significantly outperforming traditional secondary structure-based approaches. By enabling the design, folding, and evaluation of RNA sequences that fold into desired tertiary structures.

RDesign: Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design
RDesign: Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design

Yijie Zhang*, Cheng Tan*, Zhangyang Gao, Stan Z Li (* equal contribution)

The Twelfth International Conference on Learning Representations (ICLR) 2024

We proposed an RNA sequence design approach from the tertiary structure.

RDesign: Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design

Yijie Zhang*, Cheng Tan*, Zhangyang Gao, Stan Z Li (* equal contribution)

The Twelfth International Conference on Learning Representations (ICLR) 2024

We proposed an RNA sequence design approach from the tertiary structure.

Efficiently Predicting Protein Stability Changes Upon Single-point Mutation with Large Language Models
Efficiently Predicting Protein Stability Changes Upon Single-point Mutation with Large Language Models

Yijie Zhang, Zhangyang Gao, Cheng Tan, Stan Z. Li# (# corresponding author)

Arxiv 2023

Predicting protein stability changes from single-point mutations remains a significant challenge, hindered by feature extraction complexity and limited experimental data. Leveraging ESM-based large language models, we propose an efficient method integrating sequence and structural features to predict thermostability changes, supported by a carefully curated dataset to ensure fair model comparison.

Efficiently Predicting Protein Stability Changes Upon Single-point Mutation with Large Language Models

Yijie Zhang, Zhangyang Gao, Cheng Tan, Stan Z. Li# (# corresponding author)

Arxiv 2023

Predicting protein stability changes from single-point mutations remains a significant challenge, hindered by feature extraction complexity and limited experimental data. Leveraging ESM-based large language models, we propose an efficient method integrating sequence and structural features to predict thermostability changes, supported by a carefully curated dataset to ensure fair model comparison.

* denotes equal contribution
Academic Services
Workshop Reviewer: Neurips SafeGenAI Workshop 2024
Conference Reviewer: ICLR 24’; NeurIPS 24’; AISTATS 24’; ICML 25’