This thesis presents novel investigations of three common ageing phenotypes in human population studies, using microarray technology to assess ‘transcriptome-wide’ expression in whole blood to identify mechanisms and biomarkers. Muscle strength is related to frailty and is predictive of disability in older persons. I assessed the association between transcript abundance in the InCHIANTI peripheral blood samples (N=695) and muscle strength. One gene (CEBPB) passed the multiple testing criteria, and is involved in macrophage-mediated repair of damaged muscle. I extended this work with a meta-analysis of over 7,781 individuals in four collaborating cohorts; expression of over 222 genes were significantly associated with strength, less than half of which have previously been linked to muscle in the literature. CEBPB did not replicate in these younger cohorts. I then performed the first human analysis of gene expression and cognitive function (and separately with decline in cognitive ability over nine years) in the InCHIANTI cohort (N=681), and one gene was identified; CCR2, a chemokine receptor. Evidence in mice has implicated this gene in the accumulation of β-amyloid and cognitive impairment. Finally, in a collaborative project with the Framingham Heart Study I studied age-related inflammation – another hallmark of ageing - using a novel approach to ‘transcriptome-wide’ analysis; each transcript was assessed for the proportion of the association between age and interleukin-6 (IL6) that it statistically mediated. Very few of the genes associated with IL6 alone also mediated the relationship with age. Findings include; SLC4A10, the strongest mediator, not previously linked to inflammation, and interleukin-1 beta and perforin, a cytokine and cytotoxic protein, respectively. These novel analyses highlight key molecular pathways associated with age-related phenotypes in whole blood and provide links between mouse models and humans. They provide biological insight and directions for future research.