Today, cutting edge research is being conducted at the intersection of the fields of cryptography, steganography and molecular genetics. DNA has emerged as a promising new carrier medium for hidden information and DNA-based watermarks, as a special kind of DNA steganography. DNA-based watermarks are helpful tools to identify the unauthorized use of genetically modified organisms protected by patents. Furthermore, they have been recently applied to distinguish between wildtype and artificially designed genomes.
However, these watermarks must not be integrated randomly into the genome of living organisms, due to the fact that they might affect cellular functions. Artificial DNA can, for instance, alter the primary sequence and thus the function of a protein when integrated into such coding regions. To counteract these problems, several groups have developed algorithms that hide information in living organisms without affecting them. These algorithms are mainly based on the degenerative genetic code.
Mutations occurring infrequently can be detected and even be corrected by some of the published methods using parity bits or sophisticated mutation correction codes. Nevertheless, the underlying assumptions of the mutation scenarios in all existing DNA watermarking procedures are rather artificial. Mutations are treated as being uniformly distributed, and thus, these methods are not able to capture and reflect the de facto underlying biological principles of mutations, which are very specific and often accumulated. Thus, realistic mutation scenarios have to be developed for establishing practically applicable watermarking procedures for biotechnological applications.