Sheng Xu 许晟

Position
PhD Student, Biomedical Engineering
Affiliation
BEAM Lab, Fudan University & Shanghai Artificial Intelligence Laboratory
Research
AI for Molecular Biology · Computational Virology · Protein & Genome Language Models
Scholar GitHub LinkedIn
Sheng Xu
Research

Focus areas

Agent-Driven Scientific Discovery

End-to-end agentic scientific discovery for virology, spanning viral metagenomics, host prediction, and large-scale analysis.

Agentic Science Viral Metagenomics Host Prediction

Structure Prediction & Design

Comprehensive benchmarks for all-atom structure prediction. Generative approaches for protein design, antibody engineering, and molecular therapeutics.

FoldBench Antibody Design Diffusion Models
Background

Experience

Education

Fudan University
PhD in Biomedical Engineering · 2023 – present
Advisor: Prof. Siqi Sun
The Chinese University of Hong Kong
MSc in Computer Science · 2021 – 2022
Advisor: Prof. Yu Li
East China Normal University
BEng in Software Engineering · 2017 – 2021

Employment

Shanghai AI Laboratory
Research Intern · 2022 – present
Publications

Selected papers

  1. Benchmarking all-atom biomolecular structure prediction with FoldBench
    S. Xu*, Q. Feng*, L. Qiao, H. Wu, T. Shen, Y. Cheng, S. Zheng, S. Sun
    Nature Communications, 2025
    Paper Code
  2. Paper
  3. Paper Code
  4. Cryo-EM reveals mechanisms of natural RNA multivalency
    L. Wang*, J. Xie*, T. Gong*, H. Wu*, …, S. Xu*, et al.
    Science, 2025
    Paper
  5. Fast, sensitive detection of protein homologs using deep dense retrieval
    L. Hong, Z. Hu, S. Sun, …, S. Xu, et al.
    Nature Biotechnology, 2025
    Paper Code
  6. π-PrimeNovo: an accurate and efficient non-autoregressive deep learning model for de novo peptide sequencing
    X. Zhang*, T. Ling*, Z. Jin*, S. Xu*, et al.
    Nature Communications, 2025
    Paper Code
  7. Paper
  8. Beacon: Benchmark for comprehensive RNA tasks and language models
    Y. Ren, Z. Chen, L. Qiao, …, S. Xu, et al.
    NeurIPS 2024
    Paper
  9. Accurate prediction of antibody function and structure using a bio-inspired antibody language model
    H. Jing, Z. Gao, S. Xu, et al.
    Briefings in Bioinformatics, 2024
    Paper Code
  10. Paper Code
  11. Paper Code
  12. Paper
  13. AcrNET: predicting anti-CRISPR with deep learning
    Y. Li, Y. Wei, S. Xu, et al.
    Bioinformatics, 2023
    Paper

Preprints

  1. Paper
  2. SCALE: Scalable Conditional Atlas-Level Endpoint transport for virtual cell perturbation prediction
    S. Chen, L. Yu, K. Jin, S. Zhang, H. Wu, W. Huang, S. Xu, et al.
    bioRxiv, 2026
    Paper
  3. Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
    W. Xu, Y. Zhou, Y. Zhou, Q. Cao, S. Li, J. Bu, ..., S. Xu, et al.
    arXiv, 2025
    Paper
  4. Accurate de novo sequencing of the modified proteome with OmniNovo
    Y. Chen, S. Qu, Z. Gao, Y. Yang, X. Zhang, S. Xu, et al.
    arXiv, 2025
    Paper
  5. A genome-scale language model uncovers animal viruses with zoonotic potential
    S. Xu, et al.
    Submitted, 2025
  6. Paper
  7. RTM-align: an improved RNA alignment tool with enhanced short sequence performance
    Z. Qiu*, S. Xu*, J. Wei, T. Shen, S. Sun
    bioRxiv, 2024
    Paper
Get in touch

Contact

Open to discussions about agent-driven scientific discovery, virus discovery from metagenomics, host prediction, and biomolecular foundation models.

Email: shengxu@link.cuhk.edu.hk ·  ORCID: 0000-0002-6507-9122

GitHub Google Scholar LinkedIn