Knowledge Graph & Linked Data

Our research in Knowledge Graphs and Linked Data focuses on structuring and connecting heterogeneous data to enable intelligent, interoperable, and context-aware systems. We develop scalable methods for semantic modelling, reasoning, and integration that support open, dynamic, and trustworthy data ecosystems. Embracing the dataspace paradigm, we aim to facilitate flexible data collaboration while preserving autonomy, privacy, and data sovereignty. Our vision is to drive data-driven innovation by transforming fragmented information into connected, meaningful, and actionable knowledge. ​​

Current Projects: 

• An Integrated Graph Theoretical Substructure Similarities Searching Algorithm for Drug Repositioning and Off-Target Toxicity Assessments using Antimicrobial Resistance Model. Ministry of Higher Education, Malaysia – Translational Research Grant. Dec 2022 – Nov 2025. Co-PI.  We are developing a graph-based 3D substructure-similarity workflow to identify repositionable drugs from approved-drug libraries while flagging probable human off-targets, using antimicrobial resistance as the model system. Binding-site motifs are encoded as residue/atom graphs with explicit geometry and systematically interrogated across bacterial and human proteomes (PDB/AlphaFold) using tolerance-aware subgraph isomorphism. High-scoring candidates are prioritised with lightweight docking and ADMET filters, then progressed to focused experimental validation on curated AMR panels. The project will deliver a reproducible toolkit and web service that broaden discovery beyond exact-match queries while reducing computational cost and turnaround time.