Tobacco genome research is expected in the next few years to improve our knowledge about the interactions between genes involved in the formation of undesirable compounds in cigarette smoke. A close link exists between gene expression in the tobacco plant, chemical composition of raw tobacco, and combustion products in cigarette smoke. The aim of the ESTobacco project is to be complementary to other projects currently underway concerning the tobacco genome. Our strategy is to sequence only genes expressed in tobacco and not the whole genome. The size of the tobacco genome is too large to be totally sequenced (29 times more than A. thaliana). This project used three "commercial" varieties of tobacco widespread throughout the world: K326 for the flue-cured type, Burley 21 and TN86 for the Burley type. In order to obtain the major genes, the organs of the plant (seeds, roots, stem, midrib, lamina and flowers) prepared at different stages of development (germination, young seedlings, before and after topping, maturity) were used as a basis for this work. A large tobacco expressed sequence tag (EST) dataset was obtained from 11 normalized cDNA libraries comprising 56,000 clones. A DNA array designed with these sequences could allow the large-scale study of the genes expressed in tobacco. This new tool will lead to the acceleration of programs already underway concerning the origins of risks associated with tobacco and inform strategies for harm reduction. In order to encourage a wide range of initiatives on tobacco plant genetic, as with other crops, the resulting sequences obtained during the ESTobacco project are available to the worldwide scientific community through public access databases.