Xiang Ji

Xiang Ji

Auto-generated profile, based on publicly available information

Xiang Ji

School of Medicine, and Warshel Institute for Computational Biology, The Chinese University of Hong Kong, Shenzhen, Guangdong, 518172, P. R. China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 December 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
YCYigang Chen
Yigang Chen
XJXiang Ji
Xiang Ji
ZZZiyue Zhang
Ziyue Zhang
ZZZihao Zhu
Zihao Zhu
YZYuming Zhou
Yuming Zhou
HHHsien-Da Huang
Hsien-Da Huang
Deep learning-based drug-target interaction (DTI) prediction methods have demonstrated strong performance; however, real-world applicability remains constrained by limited data diversity and modeling complexity. To address these challenges, we propose SCOPE-DTI, a unified framework combining a large-scale, balanced semi-inductive human DTI dataset with advanced deep learning modeling. SCOPE-DTI is constructed from 13 public repositories and expands data volume by up to 100-fold compared to common benchmarks such as the Human dataset. The SCOPE model integrates three-dimensional protein and compound representations, graph neural networks, and bilinear attention mechanisms to effectively capture cross domain interaction patterns and outperform state-of-the-art methods across various DTI prediction tasks. Additionally, SCOPE-DTI provides a user-friendly interface and database. We further demonstrate its effectiveness by experimentally identifying anticancer targets of two bioactive natural compounds. By offering comprehensive data, advanced modeling, and accessible tools, SCOPE-DTI accelerates drug discovery research.

8 January 2026 in bioRxiv
Verified
Preprint
Open access
LTLin Tingsyuan
Lin Tingsyuan
YCYigang Chen
Yigang Chen
XLXin Li
Xin Li
XJXiang Ji
Xiang Ji
SHShenghan Huang
Shenghan Huang
HHHsien-Da Huang
Hsien-Da Huang
The photoprotective efficacy of natural skin-active complexes is well recognized, yet their ap-plication is often hindered by the challenge of deciphering their complex, multi-component, and multi-target mechanisms. To bridge the gap between established phenotypes and molecular mechanisms, we developed an AI-driven integrated platform that combines phytochemical profiling, network pharmacology, and deep learning-based target prediction with rigorous biophysical validation. We applied this platform to investigate RRBE, a bioactive complex refined from red rice bran extract. In vivo clinical studies confirmed that RRBE significantly accelerates the resolution of UV-induced erythema, while cellular and 3D tissue models demonstrated robust suppression of oxidative stress and DNA damage responses. To decode its material basis, the platform deconstructed RRBE into 10 distinct chemical modules. Leveraging our SCOPE-DTI deep learning model for global target prediction, we identified flavonoids (Module 1) and phenolic acids (Module 10) as the primary bioactive drivers relevant to photoprotection. These computational predictions were structurally supported by molecular docking and definitively validated in a physiological environment via Cellular Thermal Shift Assay-Mass Spectrometry (CETSA-MS). Mechanistically, RRBE functions through a synergistic polypharmacology: (1) Module 1 components, represented by procyanidin B2, target NDUFA7 to stabilize mitochondrial function and mitigate ROS; (2) Module 10 components, exemplified by caffeic acid, ferulic acid, and p-coumaric acid, engage FKBP11 and HSP90AA1 to regulate protein homeostasis and stress responses. This work not only deciphers the polypharmacological basis of RRBE’s photoprotective action but also validates a scalable “AI-guided, cell-validated” discovery pipeline, offering a rational paradigm for uncovering the protective benefits of complex natural extracts.