This continues what we have started to describe on November 24, when we concluded the first part of this post with, “And then the fun begins.”
Well –- looking into the international patent applications of general interest (we leave the other ones aside for this time) might not always be real fun – but it is interesting almost every time. We do not select according to rigid, quasi-algorithmic criteria; during triage, we make a preliminary selection of what we perceive as potentially at the cutting edge in applied pharmacology, drug delivery, cell and tissue technology, vaccines, and biomarkers. We also tend to preselect applications that appear not follow the most-trodden paths – i.e., those who address the less conventional molecular and pathway targets, or claim compounds that are chemically unusual, or are intended for the treatment of orphan diseases.
Now how does a life scientist and information manager analyze a patent for its novelty and inventiveness, in the substantiated and practical meaning of these terms? (Which is of course to say, not in the much looser sense intellectual property law defines them?) Many who disregard patent documents as sources of information are doing so because they despair at this point.
Indeed, what makes such a document useful for pharmaceutical R&D, business analytics, and academic science is routinely buried under a daunting amount of repetitive text, legalisms, statements that can be regarded only as blaringly obvious by anybody with some deeper knowledge in the field. More often than not, references to peer reviewed papers that are cherry-picked for providing background for the reviewers at the patent offices while not exposing any “prior art” that might compromize novelty and inventiveness (this time, in the legalistic sense of these terms).
For the analyst who wants to extract the actually useful information from a patent document that can run for several hundred pages, the challenge is to cut through this junk (and yes, from the perspective of such a person it is junk, as indispensable as these parts of the patent’s “specification” – i.e., the body text, as opposed to the claims – is for the patent attorney), and to find the “gems.” To this end the analyst will also disregard (for the moment at least) the huge amounts of chemical information that a pharmaceutical 
patent document often contains – synthesis protocols, NMR characterization data, and the like. Also, during this first pass it is usually not required to follow up with the deeper implications of Markush formulas (the generalized chemical structures with the many R’s and X/Y/Z’s in them, and the subsequent definitions of the substituents). To find out if a particular compound is actually covered, you can always circle back later.
The immediately useful information is mostly found in three sections: the first few sentences of the “Summary of the Invention,” which introduces you to the subject matter; the first few paragraphs of the “Detailed Description of the Invention;” and those parts of the “Examples” section that provide biological data in support of the invention.
The third (and last) part of this mini-series will describe how we put this extracted information in context.