PTD/14/076 - Addition of Chemical Search capabilities to the WIPO PATENTSCOPE search system

WIPO has issued this RFP (Request for Proposal) in order to obtain financial offers from the key Service providers in the Chemical Search consultancy field in response to the requirements outlined below.

Different kinds of texts are indexed in PATENTSCOPE: patent titles, abstracts, description and claims. The goal of this project is to add to PATENTSCOPE “exact” search capabilities for chemical compounds in these fields, in the sense that chemical fuzzy search, sub-compound search and Markush search are currently out of scope of this project.
The main issue to solve is recognition of the numerous different ways to refer to a given chemical compound in a patent application: accepted name, commercial name, IUPAC name, CAS name, CAS registry number, InChI, SMILES, embedded drawing, MOL file, etc. The idea is to enrich the texts indexed in PATENTSCOPE by replacing the detected compounds by their corresponding IUPAC International Chemical identifier key (InChIKey).

A first usage of these new indexation capabilities will be the implementation of a search capacity for International Nonproprietary Names (INNs) that can be represented by an InChIKey in PATENTSCOPE which may be used by the public at large (INNs are standard names of pharmaceutical drugs designated by the World Health Organization (WHO)).
Processing pipelines should be developed, taking in parameter source text strings of patent applications in HTML, localizing the references to chemical compounds (from substrings of text and embedded images), determining the compounds chemical structures and outputting the enriched text, replacing each found source text/image with the corresponding computed InChI key.
An intuitive chemical search graphical interface should be developed as an additional menu item “Chemical compounds” under PATENTSCOPE main search drop down menu item.