An algorithm to reduce multi-sample array CGH data from thousands of clones to tens or hundreds of clone regions is introduced. This reduction of the data is performed such that little information is lost. which is possible due to the high dependencies between neighboring clones. The algorithm is explained using a small example. https://www.spidertattooz.com/Varvaglione-Primitivo-di-Manduria-Papale-2020-750ml/