Gareth Williams began his research career in theoretical elementary particle physics. In 1998, he shifted his focus to bioinformatics, starting with computational approaches to protein inhibitor design. His later research concentrated on using gene expression profiles to quantitatively define phenotypes that describe both disease states and drug activities. To achieve this, he developed the Searchable Platform Independent Expression Database (SPIED) web tool (www.spied.org.uk), which integrates expression data across multiple platforms and addresses the challenge of scoring transcription profile similarity through a ranking scheme. A significant success of SPIED was the identification of robust gene expression signatures associated with neurodegenerative conditions, thereby defining a disease-associated quantitative phenotype that can serve as a valuable tool for drug discovery. More recently, his research has transitioned to epidemiology, leveraging publicly available data on dementia incidence, diagnosis, observation data and drug prescriptions to inform drug repurposing efforts.