We've reached the point at which the whole "natural" vs. "artificial" dichotomy just doesn't pass muster anymore. Advancements in synthetic biology promise to blur a line that wasn't especially crisp to begin with, especially when discussing raw DNA sequences. The SCOTUS ruling states "cDNA does not present the same obstacles to patentability as naturally occurring, isolated DNA segments...creation of a cDNA sequence from mRNA results in an exons-only molecule that is not naturally occurring." The issue here really shouldn't be whether cDNA of a patented sequence could be found in a human (that is, without influence by outside factors like viruses), but rather that the cDNA still bears exactly the same protein-coding message as the RNA transcript does, just in an edited format.
We can, of course, get more philosophical about the issue and debate what constitutes a "human" gene. At least eight percent of the sequenced human genome is made up of retroviral sequences. Some of them even code for things in active use. There have also been suspected instances of bacterial sequences jumping into the genome of their human host. I suppose each of these issues will create their own legal issues when the time comes, i.e. the next time a biomedical company gets overzealous about their intellectual property. In the meantime, this new ruling will have to suffice.
Update: I like the 2010 district court ruling. It went farther than today's SCOTUS ruling.