A technique for gene expression analysis that builds on the random sampling of cDNA libraries.
- Generate a 'sequence tag' for each mRNA in your collection. This is a characteristic 10-14bp segment that is cut from the mRNA with a type IIs restriction enzyme. Rest assured, there is a favoured type IIs restriction enzyme for this line of work.
- Join several sequence tags together to form concatemers, which are then cloned and sequenced.
- Identify the genes responsible for each sequence tag by searching databases (special sequence tag repositories are available eg. from GenBank).
This represents a massive increase in sequence throughput. The sequence tag is long enough to almost always uniquely identify an mRNA (fake proof: 4^14=bignum) if the restriction enzyme we used to select it is known; different restriction enzyme, different sequence tag - which is occasionally useful.