The term computer virus came initially from living viruses. Viruses are basically DNA (or RNA) sequences that encode for a few proteins only (those protecting them against the hostile environment and those required to replicate them in the host).
Now researchers from the University of Washington demonstrated that DNA sequences can not only be a threat to humans, animals or plants, but can work as computer viruses by encoding for malicious pieces of software - malware. When the DNA sequencers (machines that unravel the DNA sequence of small samples up to DNA from whole organisms) read that hidden attack code and compressed the sequence, the virus hacked that compression software with its buffer overflow exploit, breaking out of the program and into the memory of the computer running the software to run its own arbitrary commands.
It didn´t work that perfect yet, the researchers cheated also a little (they changed the open source code of the compressing software slightly to add a vulnerability), but still to me this a quite clever and truly pioneering piece of applied science.
For now this is of course rather theorethical, but think about sensitive DNA databases (in future our DNA is kept on file in case of diseases, etc). Criminals might want to have criminal DNA databases changed. Crop companies want their valuable intellectual property (DNA of genetically engineered plants) protected, and maybe one day data storage via DNA becomes feasible (DNA is since millions of years a biologic data storage that works very efficient).