Programme across the three institutions.
AI for Science (artificially intelligent-driven scientific research) has emerged as a global interdisciplinary frontier in artificial intelligence development. Aiming to leverage AI models for learning scientific principles, discovering scientific laws, and addressing scientific problems, it is hailed as the "fifth paradigm" of scientific research. Notable advancements have been widely recognized by academia and industry in fields such as drug discovery, materials design, and high-energy physics. This project intends to focus on drug molecular dynamics prediction and structure generation as scientific applications to investigate spatiotemporal graph neural network (STGNN) models. Centered on "spatiotemporal dynamics" and "graph (relational) structures" as core analytical objects, the model's universality stems from two fundamental laws: "The world is material, and matter is in motion" (spatiotemporal law) and "Everything is universally connected" (relational law). The specific research contents are outlined as follows:
(1) 3D Geometric Attribute Graph Modeling and Representation Learning
3D geometric attribute graph modeling is designed to convert spatially structured molecules into computerprocessable data structures. Representation learning aims to compress molecular information— including spatial geometry, topological relationships, and chemical properties— into low-dimensional, continuous, and task-agnostic embedding vectors while preserving information completeness.
(2) Mechanism-Data Dual-Driven Spatiotemporal Dynamics Prediction
Spatiotemporal dynamics prediction focuses on modeling the evolutionary laws of molecular systems across time and space, which is of great significance for tasks such as protein folding and ligand binding. This project proposes a hierarchical spatiotemporal graph neural network model based on logical and mechanistic representations to achieve accurate prediction of molecular dynamics.
(3) Geometric Diffusion Models for Molecular Structure Generation.
This research aims to develop geometric diffusion models for generating targetspecific molecular structures. Guided by conditions such as desired molecular properties or specific docking conformations, the model will generate valid and interest-aligned 3D molecular structures. Additionally, interpretable attribution methods will be employed to verify the rationality of the generated structures.
Joint Supervisory Team:
- XJTLU supervisor: Dr Yushan Pan
- Co-Supervisor: Prof. Zhijie Xu
- XJTU supervisor: Professor Pinghui Wang
- UoL supervisor: Professor Xiaowei Huang
Requirements:
A UK first-class or upper second-class honours Bachelor's degree and a UK Master's degree with Merit (or their equivalent) are required for PhD admissions. Exceptional candidates holding only a Bachelor's degree may be considered on an individual basis.
Evidence of good spoken and written English is essential. The candidate should have an IELTS (or equivalent) score of 6.5 or above, if the first language is not English.
For more information about entry requirements and admission procedures of PhD programme at XJTLU, please visit:
Programme Structure and Fees:
Doctoral students in the joint programme are registered with both XJTLU and the UoL. Upon successful completion of the programme, the students will be awarded a PhD degree from University of Liverpool. During their doctoral studies at XJTLU, students are expected to conduct research at XJTU as visiting students. Additionally, students have the opportunity to apply for a short-term research visit to the UoL if the project requires it.
The joint doctoral programme is offered on a full-time basis, with a typical duration of three to four years. The tuition fee is RMB 99,000 per year.
How to Apply:
Interested applicants are advised to email Yushan.Pan@xjtlu.edu.cn and/or phwang@mail.xjtu.edu.cn the following documents for initial review and assessment (Please include the project title in the subject line):
- CV
- Two formal reference letters
- Personal statement outlining your interest in the position
- Certificates of English language qualifications (IELTS or equivalent)
- Full academic transcripts in both Chinese and English (for international students, only the English version is required)
- Verified certificates of education qualifications in both Chinese and English (for international students, only the English version is required)
- PDF copy of Master Degree dissertation (or an equivalent writing sample) and examiners reports available
