DrugSimDB
databaseMulti-modal drug–drug similarity database: 10,367 small molecules and 238,635 significant drug pairs integrating chemical, target-sequence, functional (GO/PPI) and pathway evidence, with an interactive network explorer.
Imam Mohammad Ibn Saud Islamic University (IMSIU) · Riyadh, Saudi Arabia
We build Bayesian, network-science and AI methods that turn noisy, high-dimensional data into mechanistic insight, from drug-resistant cancer signalling and drug repositioning to trustworthy clinical AI and the Global Burden of Disease.
Methods developed for one domain, such as MCMC samplers for Bayesian networks or network-proximity scores, carry over to the next. That shared mathematical core is what links our work.
Probabilistic graphical models, MCMC samplers and causal network learning for noisy, high-dimensional data.
Modelling how signalling pathways rewire and cross-talk when tumours escape targeted therapy.
Multi-modal drug similarity, interactome proximity and combination-therapy prioritisation.
Biomarker discovery from bulk, single-cell and spatial omics with statistical and deep learning.
Explainable, federated and efficient deep learning for imaging, physiological signals and clinical data.
Population-scale burden estimation and real-time epidemic analytics.
New optimisers and loss functions, vision-language agents and physics-informed learning.
Plain-language write-ups of every method the lab has proposed, each with the math, the results and links to code.
Artificial Intelligence Review · 2026
Expert Systems with Applications, 134644 · 2026
Journal of Personalized Medicine 16 (8), 423 · 2026
Journal of Clinical Medicine 15 (14), 5651 · 2026
Our redesigned site brings together 113 publications, open-access PDFs and a methods blog with a post for every method we have proposed.
Published in Artificial Intelligence Review with collaborators at Charles Sturt University.
One of a series of Global Burden of Disease 2023 papers this year, alongside measles (Nature Medicine), heart failure (JAMA Cardiology) and elevated LDL-C (JAMA).
A DBN-GJO-PI-RNN surrogate that cuts RMSE by 38% compared with the best data-driven baseline (Polymers).
Integrated pan-cancer systems biology and structure-based simulation (Computational Biology and Chemistry).
Coupling each gene's degree of dysregulation with the cross-talk between enriched pathways shows p53 and FcεRI signalling orchestrating 87% of the perturbation map in dasatinib resistance, and nominates...
Cancer Systems Biology & Drug ResistanceA lightweight CNN diagnoses plant diseases from leaf images (82–97% accuracy across five datasets, 2.1 ms inference) and returns both automated and expert-guided treatment advice.
AI Foundations & Applied AIA web application that pairs polynomial regression with differential privacy to estimate land values from local amenities, and adds affordability, habitability and terrain insights through Google Maps.
AI Foundations & Applied AIMulti-modal drug–drug similarity database: 10,367 small molecules and 238,635 significant drug pairs integrating chemical, target-sequence, functional (GO/PPI) and pathway evidence, with an interactive network explorer.
Network-pharmacology platform to visualise and quantify how drug targets interact with the SARS-CoV-2–host interactome and to prioritise synergistic drug combinations.
Real-time visualisation and short-term forecasting of COVID-19 spread with customisable time-series models (formerly "COVID-19 TS").
Pathway Perturbation Driver discovery: learns causal Bayesian networks of a perturbed signalling pathway with Neighbourhood, Hit-and-Run and Metropolis–Hastings samplers and ranks driver genes with the p1-model....
Where our team has trained and worked
We welcome motivated PhD, MSc and undergraduate researchers, visiting scholars and collaborators in bioinformatics, Bayesian statistics, network science and AI. Our projects are computational and our collaborations are international.