What is UniChem?

UniChem is a simple, large-scale, non-redundant database of pointers between chemical structure identifiers. Its purpose is to optimise the efficiency with which structure-based hyperlinks are built and maintained between chemistry databases. This is particularly suitable for creating links ‘on the fly’ by the use of REST web services. Primarily, this service has been designed to maintain cross references between EMBL-EBI chemistry data. These include primary chemistry databases (ChEMBL and ChEBI); and secondary resources which may contain small molecule information (Expression Atlas, PDBe).

Why do we need UniChem?

When exploring small molecule information held in a particular database, a common question for many people might be “is there additional data on the same or similar molecules in another resource?”. Sometimes hyperlinks on the page can direct you to other websites and repositories where you can find this data. However, these links can quickly become out of date, or only link to a small number of other resources; often the links will not be present at all.

This course focuses on how UniChem can be used to look for such additional data. If implemented within a database, UniChem can work in the background to create and maintain these links for the benefit of all resource users. The advantage is that UniChem is free and adds no ongoing maintenance cost to the resource that employs it. If UniChem is not integrated into a site, users can still find related molecules by using the UniChem web tool directly.