PROJECT 01 · ONGOING
Genomic & epigenetic basis of neurodevelopmental and psychiatric disorders
scRNA-seqscATAC-seqBulk DNA-seqAutism & schizophrenia
Question: When during brain development, and through which regulatory DNA regions, do the genetic risk factors for autism, schizophrenia, and related conditions begin to act?
Approach: Integrating gene expression (bulk and single-cell RNA), chromatin accessibility (single-cell ATAC-seq), and bulk DNA sequencing from human brain tissue across developmental stages, in the Sanders and Rinaldi labs at Oxford's IDRM.
Status: Ongoing — contributing to lab preprints on spatial transcriptomics and spatiotemporal gene enrichment in autism and schizophrenia.
Related publications →
PROJECT 02 · ONGOING
Enhancer discovery for therapeutic targeting
Functional genomicsGene regulationTherapeutics
Question: Which enhancers — the non-coding DNA regions that switch nearby genes on and off — could be targeted to correct disease-associated gene expression in the brain?
Approach: Combining chromatin accessibility and expression data to prioritise candidate regulatory elements linked to neurodevelopmental and psychiatric disease genes.
Status: Ongoing, alongside the Therapeutic Genomics Centre's wider work on enhancer-targeted therapies at Oxford.
More about this work →
PROJECT 03 · COMPLETED
Machine learning for drug–drug interaction & disease-association prediction
Neural networksMatrix factorizationBioinformatics
Question: Can integrated similarity measures between drugs — chemical, target, and side-effect profiles — predict interactions and disease associations that haven't yet been observed clinically?
Approach: Built neural-network (NDD) and integrated similarity-constrained matrix factorization (ISCMF) models, plus a semi-supervised graph-cut method for drug–disease association prediction.
Status: Completed; published in Scientific Reports and Network Modeling Analysis in Health Informatics and Bioinformatics.
See the code →
PROJECT 04 · COMPLETED (PHD)
AI-driven insight into health data science & precision medicine education
Educational data miningLearning analyticsPrecision medicine
Question: How do students learn precision medicine and health data science concepts, and how can that process be better supported and predicted?
Approach: Applied AI and educational data mining to student self-reports and performance data as part of the MRC-funded Precision Medicine Doctoral Training Programme, supervised by Dr Kobi Gal, Dr Areti Manataki, and Dr Michael Gallagher.
Status: Completed — PhD, University of Edinburgh. Findings published in Studies in Health Technology and Informatics.
Related publications →