The aim of genetic studies in migraine is to identify key proteins in order to better understand the molecular mechanisms of this frequent but still incompletely understood condition. This review describes the current knowledge in the field of migraine genetics. Migraine genes have been, and still are, difficult to identify. The more common varieties of migraine are characterized by a high prevalence in the general population, and a high phenotypic variability. In the absence of any objective diagnosis marker, the status for genetic studies is established only clinically. The first breakthrough was permitted by the study of familial hemiplegic migraine, a variety of migraine with motor aura. This rare condition has a monogenic, autosomal dominant mode of inheritance, thus enabling genetic studies. The three first genes, identified from 1996 to 2005, all encode ion-channel transporters: a neuronal calcium channel (CACNA1A, FHM1), a glial sodium/potassium pump (ATP1A2, FHM2) and a neuronal sodium channel (SCN1A, FHM3). Study of cellular and animal models have shown that mutations in CACNA1A and ATP1A2 facilitated the initiation of cortical spreading depression waves, the mechanism underlying the migraine aura, and most likely increased neuronal excitability with an excess of glutamatergic neurotransmission. In 2012, PRRT2 has been identified as the fourth FHM gene, and encodes an axonal protein associated to the exocytosis complex. In the 1990s, family and twin studies showed that the more common varieties of migraine (migraine without aura and migraine with typical aura) were polygenic, with an overall heritability nearing 50 %. These genetic factors interact with environmental factors. The initial attempts to identify migraine genes by candidate gene approaches or by linkage studies were deceiving. Since 2010, three large genome-wide association studies (GWAS) have identified six genetic variants associated with migraine. Each variant has only a modest contribution to the overall genetic risk of migraine, suggesting a marked genetic heterogeneity. Three of the migraine-associated variants affect genes involved in glutamate homeostasis. Another variant concerns a gene encoding a protein implicated in nociception. Three of the four polymorphisms are associated both with migraine without aura and migraine with aura, supporting the existence of molecular mechanisms shared by all varieties of migraine. The vast majority of the migraine genes are still to be identified. Future researches will rely on new GWAS on larger cohorts of patients and controls, with a better phenotypic assessment, and on extensive sequencing.