Tobacco is an allotetraploid (2n = 4x = 48) and most likely derived from a tetraploidization event involving the diploid species Nicotiana sylvestris (S-genome, 2n = 24) and Nicotiana tomentosiformis (T-genome, 2n = 24). Two full length cDNA libraries were constructed using various tissues of N. tomentosiformis and N. sylvestris. Sequencing was performed and over 20000 EST were obtained. Combining 350000 Nicotiana tabacum ESTs from NCBI, all ESTs were assembled using the Program CAP3, resulting in 34450 contigs and 123511 singletons (157961 unigenes). The transcripts from resident T- and S-genomes in the allotetraploid nucleus are more closely related to their diploid parental homologs than they are to each other. Thus, combining two diploid EST sequences with the ESTs of tetraploid species does not introduce appreciable amounts of additional sequence divergence. The UniProt BLASTX hits, GO annotation, and Pfam were analyzed. From the EST alignments, we inferred 127787 single nucleotide polymorphisms (SNPs) and 11869 Microsatellites (SSRs). The Nicotiana EST assembly along with SNPs and SSRs will provide a foundation for future investigation of tobacco functional and evolutionary genomics using both long and short oligonucleotide microarrays. The newly identified SNPs and SSRs in our study can be used to generate a genetic map and locate genes of economically important traits within the tobacco genome. These markers will pave the way for successful application of Marker-assisted-selection (MAS) in breeding programmes.