R&D in chemistry and pharmacy requires fast and complete information on properties of relevant substances and on any existing patent protection. Patent information, if obtained early, may help to avoid double or unnecessary development and to purposefully use precious resources.
A substance identity and properties are defined by its molecular structure. Consequently, information on biological, pharmaceutical, and all other properties of a substance, on synthesis, and any existing patent protection, is stored in databases, together with the molecular structure.
Substance information is therefore best retrieved using the molecular structure. Alternatives such as searching a name or even a CAS registry number will often result in incomplete retrieval. For example, complete information on the active agent fluoxetine, its stereo isomers, salts, and so on is best found by searching the following structure:
Similar substances have similar properties; therefore a patent search often has to cover a whole group of similar substances. Often, all structures that share a common core structure are of interest. A substructure search on the example structure above will retrieve all known substances that share the fluoxetine core structure and bear any further substituents.
Many not yet synthesized compounds are nevertheless protected by patent law, if claimed in a patent document in a Markush structure, a structure containing generic variable symbols. These may be compounds bearing substituents in variable positions, having variable ring sizes or chain lengths, variable atom types, and many other variations. Such structures cannot be retrieved in the CAS Registry file or in the World Patents Index. Moreover, well-known compounds may be hidden in patents in a Markush structure, such as fluoxetine in the following two structures.
Markush structure searches can be performed in databases such as MARPAT on STN, or MMS on Questel.