Patents - January 26, 2018

Biosoftware eligibility at USPTO: mere computations or improvement in technology?

Bioinformatics has become an increasingly important tool in understanding biological processes and promoting and supporting progresses in molecular biology and many other areas of biology.

The related software involves massively complex calculations that predict the actions – folding, bonding, membrane translocation – of molecules. These software products are incredibly important in the biotech field – for example, drug design or protein folding.

And yet, in the wake of the Alice decision, we are observing a constant barrage of rejections from the USPTO under 35 USC 101 claiming that since bioinformatics methods are computational in nature and do not “improve the computer”, they are not eligible for patentability even if they are novel and non-obvious. A common point made in these 35 USC 101 rejections is that the software does not “improve the computer or computer technology.” However, it is important to remember that Alice also notes that method steps that “effect an improvement in any other technology or technical field” can satisfy the “significantly more” criteria for patentable subject matter. There is no apparent basis in Alice or other court decisions that would support a conclusion that if a method does not improve the computer” is not eligible for patent protection if it provides “an improvement in other technology or technical field”. Clearly, bioinformatics software improves the technical field of molecular biology.

Still, rejections of bioinformatics might be due to the way claims are written (of course) as well as due to the uncertainty or a narrow reading of what “effect and improvement in any other technology or technical field” means. Does it require a hardware component, like the thermocouple of Diamond v. Diehr? This seems unlikely, because the court in Alice didn’t say “effect an improvement in a machine of any other technology”. And wouldn’t an improvement to a “technological field” be broader than that?

Therefore, we draft claims in which the steps of an inventive biosoftware method are characterized by transformative technical features so that the steps and related combination are tailored to the specific features of the biological environment that is simulated, such as molecule modeling, changes in the related residues or moieties in view of environmental conditions, molecule interactions and the like.

However, what we sadly observe is that some Examiners of the USPTO look at bioinformatics software and can’t view the claims as being anything more than being “directed to” computations, data manipulations, and math – no matter how the claims are written.

The Examiner dismisses technical features such as representation of spatial locations of a biomolecule in a biological environment or representation of changes in such location due to molecule interactions with the environment, and just interprets those claims as processing information and converting one form of numerical representation into another by organizing information through “mathematical concepts such as mathematical algorithms, mathematical relationships, mathematical formulas, and calculations.

Such an approach de facto ignores looking at the claims “as a whole” and misses seeing the technical features of the claims – method steps that detail bioinformatics computations that specifically produce a result useful only to predicting the behavior of molecules – that lead the claim to actually being directed to an improvement to a technical field.

Additionally, the USPTO puts this software claims into a catch-22 scenario whereby the technical features included in the steps of the method claims are ignored and adding steps of using the results of the software (such as drug production) is dismissed as merely stating “apply it” to method, and the omission of such a step is pointed out as lacking some sort of “transformation” in the technical field.

If only the USPTO examiners would take a step back and look at how integrated claims written as indicated above and considered as a whole, are in the technical fields of bioinformatics and molecular biology, then they would agree that this type of software should be observed as satisfying the two-part test of Alice. Even if the Examiners still disagree, they need to point to the court cases that support their position, since the USPTO does not have the power to create new forms of judicial exception to patent eligibility.

Thankfully, there has been a recent indication that the USPTO might be going in this direction. In late November of 2017, there was a USPTO PTAB appeal decision (2017-003914; for application 13/822,231) where the Board stated that the examiners “”must be careful to avoid oversimplifying the claims” by looking at them generally and failing to account for the specific requirements of the claims” (quoting McRO). They also pointed out that the bioinformatics software claims were “focused on a specific articulated improvement in protein modeling” and were “distinguishable from those in Alice and related cases, in which the computer-implemented method was responsible for merely “organizing [existing] information into a new form” or carrying out a previously-known and fundamental economic practice.” Hopefully, this type of analysis will extend to other applications as well.

The IP portfolio of an entire sector of industry and R/D is being wiped out for no valid reason other than the grey morass that was left in the wake of Alice.

As USPTO Director Andre Iancu wrote in his paper “Machines and Transformations: The Past, Present, and Future Patentability of Software” (concerning the Bilski case), “Whatever the merits of various arguments regarding the patentability of business methods, the patentability of software processes, which are at the heart of innovation in our Information Age, at least deserves separate consideration.” And, likewise, we hope that the exclusion of economics-based software in Alice does not impede the patentability of technologically based software methods.