Central crop databases or registries are important tools to enhance the use and conservation of plant genetic resources. In 2008–2009 a group of Centers from the Consultative Group on International Agricultural Research worked together on the development of central crop registries for eight different crops. The International Rice Research Institute and the International Maize and Wheat Improvement Center led the development of the crop registries for rice and wheat, respectively. The registries were to compile data across collections and add value to these datasets by assessing the similarity of the accessions from those collections. We describe in detail the methodology developed for the rice and wheat registries. This methodology mainly followed an algorithmic approach to assess the correspondence between pairs of accessions. Accessions which shared a common origin were placed together in similarity groups. Using these groups the similarity of accessions in and among collections was further analysed