Placing Molecular Compounds in the Public Domain

-- By TheodoreSmith - 18 Oct 2008

Table of Contents

Introduction

Chemical compounds are protected under US patent law as "compositions of matter." Although an inventor seeking a compound patent must show a use for the claimed composition, the same patent may ordinarily be defeated by prior art showing only the structure of the molecule and "enablement," a means of successfully building or synthesizing the chemical; no showing of usefulness is required. As advances in the material sciences refine techniques for the atomic level manipulation of matter, we are rapidly nearing the point where enablement becomes trivial; any organic molecule may be constructed from its constituent atoms or pre-synthesized building blocks.

Once this threshold is reached, new chemical compositions will become far easier to push into the public domain. Even if such novel compounds do not become generally unpatentable under the doctrine of obviousness, it will become possible to render them unpatentable simply by publishing a sufficiently detailed map of the molecules structure and a simple set of instructions enabling construction. The low cost of publishing over the internet paired with the plausibility of algorithmic methods of generating molecular permutations and enablement steps almost guarantee the eventual construction of a wiki-style database of molecular permutations. This database, showing structure and enablement steps for a wide swath of potential compounds, would have the effect of rendering unpatentable every molecule appearing within.

Enablement and STMs

Advances in the atomic level manipulation of matter have brought modern science to within striking distance of the ability to manually construct novel molecules from their constituent components. In a 2002 paper, Hla and Reider detail the ways in which technicians may manipulate a scanning tunneling microscope (STM) to sever and reform atomic bonds, reposition atoms, and manipulate molecular structures to form novel compounds. Although this technology is still in its infancy, the manual construction of molecular compounds is undeniably possible, and one day may become trivial with further advances in equipment and scientific technique. In the last 19 years, the state of the art in atomic manipulation has moved from the painstaking repositioning of Xenon molecules, to the breaking and reforming of bonds within a molecule itself; it is merely a matter of time before the construction of complex molecular compounds becomes scientific reality.

Strange Properties of Molecules

Once the manual construction of any sufficiently described molecule becomes technically feasible, enablement is almost certain to become trivial to a person skilled in the technical art. Computer programs that permute and diagram extant molecular structures are scientific reality; an algorithm capable of generating build routines would likely be even simpler to develop.

Trivial enablement, by itself, is no block to patentability. Many simple mechanical devices are trivial to enable. Molecular structures, however, have the additional property of having a form consisting of a collection of discrete and finite components. The structure of simple molecule, such as H₂O, can be described in detail simply by extrapolating from the basis of its chemical formula. More complex chemical forms have many more possible structural arrangements, and are more difficult to describe; however, techniques for permuting the possible structures of these complex molecules have been developing in the prior art for some time.

The combination of trivial enablement and a finite structure would place chemical compounds within a unique class respective to patent law. A full description and build routine of a novel compound could be generated from a small amount of descriptive data, in some cases as little as a name.

The Internet Is for Publishing

Legally, the trivialization of the enablement of molecular compounds is likely to cause the most disruption within the patent doctrine of novelty. Under novelty rules, a patent may be invalidated by published prior art that both describes and enables the claims of the patent. In the field of chemistry, the creation of invalidating prior art has historically been expensive; the synthesis of a novel chemical compound is research intensive, and the cost to physically publish the information is non-trivial.

Historically, we may look at this expense as having two important effects. First, it sets the cost of generating invalidating prior art close to the cost of the research necessary to actually file a patent. Second, it makes the cost of filing a patent cheaper relative to the amount of capital spent on research and development. These ratios have encouraged the development of patents over the generation of public prior art; they have decreased the marginal cost of seeking a patent over simply publishing information into the public domain.

With the trivialization of the enablement step and the advent of near-zero-cost publishing on the internet, both these ratios shift substantially. A public internet wiki with computationally inexpensive algorithms for permuting and enabling molecular structures drops the cost of publishing invalidating prior art to nearly nothing. The invalidating enablement steps and published description that would have taken a chemist years to produce may be generated and published in the US for fractions of a cent by anyone on earth; a bored high school student playing around with chemical structures on such a wiki during class could produce and publish invalidating prior art covering thousands of molecular permutations. Although the brute force publishing of all possible permutations of all feasible molecular structures is likely impossible, and the pharmaceutical companies will likely retain some unexplored patent space in which to work, the balance between between the public and the patentable will have forever changed.

Conclusion

Although a precise legal and technical investigation remains outside the scope of this paper, it is important to note that this analysis rests not on the details of the implementation or technologies used, but on the fundamental characteristics of molecular compounds, the internet, and patent law. Similarly, it should be recognized that this change is not manifestly positive or negative. Although an economic or moral analysis of the shift may be useful in providing context, such judgments have no effect on the inevitability of the change; the cost efficiencies of internet publication and cheap computational power all but guarantee the eventual outcome.

I think this is a superb example of thinking realistically over the horizon. It would not have been sensible in 1,000 words--while introducing concepts abtruse for many (which in my view you do with clarity and economy)--to have imagined the possible legislative responses likely from an industry with so much to lose from loss of patentability. But further meditation on that line might be fruitful.

-- EbenMoglen - 15 Nov 2008