In a number of human diseases, including depression, interactions between genetic and environmental factors have been identified in the absence of direct genotype-disorder associations. The lack of genes with major direct pathogenic effect suggests that genotype-specific vulnerabilities are balanced by adaptive advantages and implies aetiological heterogeneity. A model of depression is proposed that incorporates the interacting genetic and environmental factors over the life course and provides an explanatory framework for the heterogeneous aetiology of depression. Early environmental influences act on the genome to shape the adaptability to environmental changes in later life. The possibility is explored that genotype- and epigenotype-related traits can be harnessed to develop personalized therapeutic interventions. As diagnosis of depression alone is a weak predictor of response to specific treatments, aetiological subtypes can be used to inform the choice between treatments. As a specific application of this notion, a hypothesis is proposed regarding relative responsiveness of aetiological subtypes of depression to psychological treatment and antidepressant medication. Other testable predictions are likely to emerge from the general framework of interacting genetic, epigenetic and environmental mechanisms in depression.