In the last decade, progress made in genetics is questioning the current implicit nosological model in the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR) and the International Classification of Diseases, tenth revision. Both the categorical nature and the comorbidity detected on applying diagnostic criteria become unsustainable in the light of the genetic architecture that is emerging from studies being conducted on the genetics of mental disorders. The classical paradigms -one gene for one disease- or even a specific distinctive genetic pattern for each condition, are concepts restricted to specific cases. In this review the objective is to describe the current scenario that has arisen following the latest advances in genetics. The lines of work being traced by research both in the present and in the near future include: the identification of variations in the number of copies (both frequent and rare), indiscriminately linked to different disorders; the concurrence of multiple variants for a single disorder; the double hit phenomenon; and epigenetic modulation. The new version of the DSM, fully aware of the deficiencies in the current model, will mark a turning point that, while somewhat timid, is decidedly oriented towards incorporating a dimensional conception of mental disorders.