Compound relation and chemical classification
Finding and extracting relationships
Using natural language and semantic syntax processing algorithms, ChemAnalyser modules are able to find and extract interesting relationships between chemical terms – e.g. chemical compound names, chemical substance classes, CAS numbers, patent classifications etc. – and other knowledge domains in any kind of digital documents. These comprehensive extraction modules based on a smart programming language and extensive syntax knowledge databases are covering the following relationships:
- utility of compounds
- metabolism of compounds
- structure-activity-relationships (SAR)
- structure-property-relationships (SAR)
Most of these relations are extracted and stored in dedicated NO-SQL databases. The process is fully automatic, reliable, and extremely fast. As the majority of solid scientific data can be found in tables and other figures, our smart software modules are programmed to extract data also from such sources with great precision.
Please do not hesitate to ask for a customized search user interface that satisfies your special needs.
What kind of data is extracted?
The modules,are able to find and extract relevant chemical compound data from all known digital data sources. For example, you can easily find which documents contain and describe a chemical compound for the very first time; which physico-chemical properties have been published about a special chemical compound; which proteins are inhibited by a special chemical compound or molecule structure or which diseases may eventually be treated with a special molecule or a chemical compound mixture. Such and more relationships can be found fast and reliable.
Comprehensive and competitive classification
Our chemical classification system recognizes chemical terms like IUPAC, CAS, drug names, trade names, InChI keys and trivial names correctly and fast, and subsequently classifies them accordingly. The huge amount of chemical entities – available in all kinds of documents – are stored automatically in underlying databases that can be used by our clients for sophisticated and extensive search strategies.