Update: This week, Ancestry.com responded to the lawsuit from 23andMe by filing a motion to dismiss with a California federal court. The filing indicated the 23andMe patent consisted of "abstract and non-inventive steps" of collecting two DNA samples and then comparing them to find a correlation based on phenomena that occur naturally.
The dismissal motion references two Supreme Court precedent rulings, one of which JN Murphy correctly guessed (below). This one is Alice Corp. versus CLS Bank International (2014) and is the "do it on a computer" argument. It had to do with an electronic escrow service for financial transactions, but rather than specifics of the computer-based service, the patent dealt with the idea behind the service, which evidently wasn't any more remarkable in concept than doing it with a paper ledger.
SCOTUS was unanimous in their opinion, written by Justice Clarence Thomas, that an abstract idea could not be patented just because it is implemented on a computer. Escrow is not an invention, and simply using a computer to manage it doesn't make it eligible for patent protection.
The second cited precedent is more interesting, and one I didn't see coming. This one is the March 2012 Supreme Court decision in the case of Mayo Collaborative Services (Mayo Clinic Medical Services) versus Prometheus Laboratories, Inc. This one involved a class of drugs used in treating certain autoimmune diseases. The drugs are metabolized differently by different people, so there's no one "correct" dosage. A Montreal hospital identified the threshold level, and filed patents to use that threshold as the start of steps to determine the most effective dosage for a patient. Prometheus Labs owned the patent, and up until 2004 the Mayo Clinic bought and used their "kits" for treatment. In June 2004 Mayo started offering its own diagnostic tests for the drug therapies, and Prometheus sued.
SCOTUS ruled unanimously that giving a drug to a patient, measuring the naturally-produced metabolites resulting from doing so, and then deciding whether to increase or decrease the dosage of the drug was not patentable subject matter. Justice Breyer wrote: "Because methods for making such determinations were well known in the art, this step simply tells doctors to engage in well-understood, routine, conventional activity previously engaged in by scientists in the field. Such activity is normally not sufficient to transform an unpatentable law of nature into a patent-eligible application of such a law."
Taking the two precedents cited by Ancestry, it seems their platform is that neither natural phenomena nor abstract ideas are patentable unless there is a new, additional, inventive concept involved. In other words, escrow had been done for decades before the first computer, and physicians had long adjusted drug dosages based on patient response. Ergo, with patience, you could write down the allele values resulting from the DNA tests of two individuals and compare them manually, side-by-side, for relatedness. To remind: 23andMe's infringement suit wasn't about patented testing technologies; it was about comparing DNA SNP information of two people using data stored in a database.
Maybe we'll end up with the "do it on a computer" argument, and the "DNA is DNA" argument.
Ancestry also claimed the false-advertising lawsuit--filed simultaneously but separately by 23andMe--was invalid because 23andMe is a competitor, not a customer who made decisions or purchases based on such claims. In the motion, Ancestry's attorneys write: "Not only does 23andMe fail to allege any facts establishing actual reliance on Ancestry's allegedly false and/or misleading statements, but the relief that 23andMe seeks is prohibited by well-established precedent." An allegation was that Ancestry runs "perpetual sale" by never selling the purported $99 test for $99.
Another point of contention that, to me, Ancestry deflects less well is the claim that it tests "5x more regions than other DNA tests." In this week's motion, Ancestry responds that the statement is a general one, not singling out or pointing to 23andMe. The motion also notes that 23andMe doesn't establish how many "regions" Ancestry tests in comparison to other competitors.
In its May filing, 23andMe had also claimed the "Ancestry" trademark was invalid because it is too generic. Evidently Ancestry responded to this as well, but the argument is redacted in publicly available documents.
This will no doubt drag on for many months unless one party blinks and decides to settle. But some of the issues involved are fundamental to our very notion of genetic genealogy, and worth watching for future developments.
See also: "23andMe DNA Patent Invalid Under Mayo, Ancestry.com Says," by Dani Kass; Law360.com, a LexisNexis Company.