I received my B.S. in Biological Science from Xinjiang University in June 2026, with an overall GPA of 4.30, ranking in the top 1%.
I love biology — more is different.
My academic journey includes enriching experiences at Shanghai Jiao Tong University and Peking University through summer programs.
Research Interest: AI for Synthetic Biology, bio-inspired algorithm optimization, multi-scale coupling simulation of large biomolecules (cross-system / cross-method), and trustworthy scientific AI agents.
I am actively seeking academic collaborations, particularly in AI for Science, AI Infrastructure, accelerated quantum chemistry computing, deep learning potentials, and scaling computation & model training. Please feel free to contact me via email.
Email: Zhuangziyan0623@outlook.com
Email / CV / Github / ORCID / Google Scholar
* denotes corresponding author.
An autonomous scientific intelligence platform that integrates a cross-disciplinary knowledge graph, a multi-agent reasoning system, and hierarchical scientific skills (protein design, molecular simulation, metabolic network analysis) into a closed Think→Act→Observe→Review→Re-decide loop. It enables cross-scale intelligent bioengineering — from catalytic mechanism reasoning and enzyme active-site design to whole-cell metabolic network optimization.
A production-ready, end-to-end pipeline that transforms academic PDF documents (including figures and diagrams) into a queryable knowledge graph with GraphRAG. It leverages Vision Language Models (VLMs) to extract and describe figures, and organizes each execution as an isolated run with logs, prompts, and outputs for reproducibility.
A MATLAB-based toolbox developed for comprehensive protein sequence analysis. Key features include Smith-Waterman alignment, motif discovery, and neighbor-joining phylogenetic analysis. It provides integrated visualization to streamline the workflow from raw sequence to structural insights.
A mathematical model focusing on Diffusion-Adsorption-Uptake (DAU) mechanisms. This project facilitates the efficient application of predictive values obtained from wet lab experiments to control morphogen distribution patterns.
A Deep Learning framework utilizing Temporal Convolutional Networks (TCN) and Attention mechanisms to predict pest resistance evolution, bridging mathematical models with biological pest control strategies.
Volunteered for Zhejiang University's China Temple Ancient Ginkgo Leaf Collection Initiative, performing standardized sampling of ancient ginkgo trees. Specimen data has been archived in the Zhejiang University Herbarium and incorporated into the GinkgoDB comprehensive ginkgo database, providing important specimens and data for ancient ginkgo germplasm conservation and genetic lineage research.
Participated in the summer school at the Institute of Natural Sciences, Shanghai Jiao Tong University, focusing on the application of artificial intelligence in bioengineering. Learned cutting-edge techniques including AI-driven protein design and computational biology workflows.
Participated in the summer school at the Center for Life Science (CLS), Peking University, exploring quantitative methods in biological research, including mathematical modeling of biological systems and data-driven experimental design.
Participated in field research in the Northern Tianshan Mountains, using infrared camera technology to record and analyze the behavior and distribution of local wild animals. The visual above shows representative footage captured during the internship.