The sweet, edible almonds we enjoy today are a far cry from their bitter wild ancestors thanks to a point mutation in the genes, according to a new study, which presents the completed almond reference genome. It's thought that the initial domestication of the almond tree occurred in the Near East sometime during the first half of the Holocene, a hypothesis supported by early archaeological evidence of the nut in both ancient Egypt and Greece. Wild almond species accumulate the bitter and toxic cyanogenic glycoside amygdalin, and research has suggested that initial domestication of the almond was enabled by the selection of sweet, edible kernel genotypes, which originated within the inedible wild genotypes. Since its initial domestication, the almond has become one of the most widespread tree nut species on Earth. Despite the distribution and economic importance of the almond, however, a detailed understanding of the plant's genome has lagged behind that other species in the Rosaceae family of plants, and the nature of the gene that allowed for edible, sweet kernel nuts has remained elusive. Here, Raquel Sánchez-Pérez and colleagues present the completed reference genome of the almond (Prunus amygdalus). What's more, the authors used the assembled sequence to reveal the genetic differences between toxic, bitter almonds and their sweet counterparts. Sánchez-Pérez et al. discovered a cluster of transcription factors associated with sweet kernel genotypes. Among these, bHLH2 was revealed to be involved with the regulation of the biosynthetic pathway for the production of the toxic compound amygdalin. According to the results, a mutation in bHLH2 prevents amygdalin production, resulting in the sweet almond genotype, which was actively selected for during domestication.