Dalke Scientific Software: More science. Less time. Products
[ previous | newer ]     /home/writings/diary/archive/2020/09/11/similarity_principle_in_legal_cases

Similarity principle in legal cases

Read enough cheminformatics papers and you'll eventually come across the a variation of the similarity principle or similar property principle or molecular similarity principle. Here's a 2020 example from the JCIM paper Assessing the Performance of Mixed Strategies To Combine Lipophilic Molecular Similarity and Docking in Virtual Screening:

Under the framework of the similarity principle property,5 which assumes that similar compounds should have similar properties, a plethora of methods have been developed to disclose structure-activity relationships, derive pharmacophores to rationalize the biological activity, and perform similarity measurements in the search of novel chemical scaffolds.
and here's the only one in J. Cheminformatics, titled Optimal assignment methods for ligand-based virtual screening by Jahn, Hinselmann, Fechner and Zell (2009):
The underlying assumption of ligand-based approach is that similar structures have similar biological activity. Maggiora and Johnson [8] introduced the similarity-property principle that implies that the chemical similarity can be related to the biological activity of structures.
In both cases that citation is to Johnson and Maggiora's (ed.) seminal work Concepts and Applications of Molecular Similarity.

That book is clear that they are not making new observations about the association between molecular similarity and property similarity. Indeed, the first sentence of the preface is Applications that make use, either explicitly or implicitly, of the concept of molecular similarity in chemistry are numerous, and indeed lie at the heart of a significant body of chemical research.

The book is focused on the scientific applicability of molecular and chemical similarity (it distinguishes between the two), with an emphasis on the types of molecular similarity which can be automated, while the cognitive algorithms by which medicinal chemists perceive similarity are largely unknown. (as Maggiora, Vogt, Stumpfe, and Bajorath comment in Molecular Similarity in Medicinal Chemistry (2014)).

Legal citations of the similarity principle

A point of my previous few essays was to add to the historical discussion in Johnson and Maggiora's book, with one essay on patents and the other on drug analogue laws. I'll end my series of legal quotes with a few citations using the similarity principle, drawn from US case law from before 1980, so well before Johnson and Maggiora. In the following, the bold emphasis is mine.

in re Hoch

It appears in the US the similarity principle is referred to as in re Hoch because of the Application of Paul E. Hoch, 428 F.2d 1341 (C.C.P.A. 1970):

Such actual differences in properties are required to overcome a prima facie case of obviousness because the prima facie case, at least to a major extent, is based on the expectation that compounds which are very similar in structure will have similar properties.

References to in re Hoch

This decision and summary is cited often in later judgments, as the following few examples show:

The Papesch Obviousness Regime

I found an essay titled Post-KSR Chemical Obviousness in Light of Pfizer v. Apotex by Harold C. Wegner, dated June 13, 2007. Its second paragraph is:

The law of obviousness of a chemical compound evolved from late nineteenth century structural obviousness case law that became highly defined in the 1940's and was modified in the early 1960's to provide a balanced two part analysis: The Examiner has the initial burden to establish prima facie or structural obviousness, but where this is established, it is up to the patent applicant to present rebuttal evidence demonstrating actual differences between the claimed compound and the prior art. See § II, The Papesch Obviousness Regime.
That section of the essay is:

After a generation of uncertainty and then several years of challenges, Papesch3 became the core foundation for modern chemical obviousness practice, seemingly untouched by KSR.

Under the Papesch regime, obviousness of a new chemical compound proceeds through two stages. First, is there a prior art compound sufficiently close in structure to the claimed compound to suggest that the claimed structure would have the same properties? If the answer is negative, then the inquiry is completed: There is no obviousness for such a compound. But, as is particularly the case in a mature field, there is a high degree of predictability of properties of a compound keyed to structure, such that the disclosure of a prior art compound suggests that the claimed compound can and should be synthesized to achieve like results. Here, the claimed compound is prima facie obvious based upon the concept of structural obviousness. …


Which leads us to the Application of Viktor Papesch, 315 F.2d 381 (C.C.P.A. 1963)

The examiner's statements about the existence of a family of other properties common to the claimed compounds and the compound of the prior art finds support in the record, however, only on the basis of assumptions in turn based on assumed homology. In his answer, the examiner cited a new reference, Wertheim Textbook of Organic Chemistry (2d Ed.), page 37 (1945). From it he quoted the following statement about members of any one homologous series:

"These compounds have similar chemical traits, because their structures are closely related; therefore we can learn the chemistry of the entire group with no more effort than would otherwise be required to study a single compound."

… Wertheim, at the point from which the above quotation was taken, was discussing the same methane series of hydrocarbons as Fieser and Fieser. The examiner stopped his quotation just before the sentence reading, Of course we may anticipate certain exceptions to this general rule, but such exceptions will make very little trouble. We are not here dealing with the methane series or with the type of homology which it illustrates.

We have had sufficient contact with homology on this court to agree with the examiner that such similarity in structure as exists here probably indicates similarity in some undisclosed properties; but we are past giving too much legal significance to the bare term homolog, even where there is an admission of homology, as there appears to be here. The term is often used loosely. So far as we know, the assumed similarities referred to by the examiner are of little or no practical or commercial significance. Certainly he has pointed to none. On the other hand, the proven dissimilarity is a matter of pharmacological significance, on which the examiner would be quite willing to grant a patent if the invention were claimed as a process. As to the other properties, nothing in the record gives us any information, not even the Robins et al. reference.

Andrew Dalke is an independent consultant focusing on software development for computational chemistry and biology. Need contract programming, help, or training? Contact me

Copyright © 2001-2020 Andrew Dalke Scientific AB