23andme is actually doing original research. It scares me a little bit that I did it though because 1) I am afraid of genetic discrimination even though some states have made this illegal. 2) I found out genetic disorders that I did not know I had. Fortunately none are horrendously serious but I found out what will probably kill me at some point like it killed some of my grandparents.
For research, they ask as many survey questions as you are willing to answer - anything from "Do you have psoriasis?" to "Do you cry easily?" and many many more. They are then able to correlate answers to these questions with your genetic makeup. The final result should give additional information when aggregated among all their customers.
Beyond that, I assume that many genes aren't patented just like many business processes are not patented. The person that first discovered the gene didn't go through the time and expense to patent it (for whatever reason) and because there was prior art, it's not a good target for other people to patent either.
I have no idea about estimating the actual impact on spending.
Part of me thinks "of course genes can't be patented, no one invented them", and then part of me thinks "using the patent system to protect gene isolation is an interesting legal hack because we don't have a better way to protect the investment involved with gene discovery." And surely it's got to cost something to figure out which genes are interesting to test for, which is the real value in the gene test, right?
But if that's the case, how is anything on the 23andme test cheap? Publicly funded research? Benevolent release of research from private labs? Old research that no one thought to patent? Much more reasonable licensing terms? Oversight by the current patent holder?
How realistic is the specter of "No one will do gene isolation after this decision?"