Difference between revisions of "PSICQUIC Web Services for iRefIndex"

From irefindex
(No difference)

Revision as of 12:05, 16 September 2011

These web services been been officially released and are further described in a Nature Methods article available here

"PSICQUIC" is pronounced like the English word "psychic" as in the ability to perceive information hidden from the normal senses through extrasensory perception.

A number of use cases are envisioned for these services. For example, once implemented by multiple interaction databases, they would allow a user to make a query for information at one database web-site and then have that query replicated at all other interaction databases (that have implemented the service interface). The user could then receive a compiled list of results that match their query regardless of the database that contains the information.

Note that only the first 15 columns of the iRefIndex MITAB files are returned from this service. The full data set is available via FTP download here and via iRefWeb.

WSDL locations:

Version information:

Available services:

All active services:

REST service samples:

More details about the project:


The format and content of the interaction data returned by these services is described at


Changes from last version

There are major changes between iRefindex psicquic service version 1.1.5 and version 1.1.6. In 1.1.5 the results returned were interaction centric (one line in results representing one binary interaction or a complex membership). From 1.1.6 onwards the results will be experimental centric (one line per evidence per source). The new extended PSI MI-TAB (version 2.6) is used in the implementation but, only the first 15 columns will be available at the moment in order to maintain cross compatibility. For further details on the returned results please refer the following WIKI page;

The second major change is the usage of UniProt and RefSeq identifiers as uidA and uidB when ever that is possible.

Users are advised that there are differences between our implementation of these services and that of other databases. Most notably,

  1. interactions between 3 or more proteins (n-ary data) is represented in the PSI-MITAB files using a bi-partite model (one of the first two columns is used to represent the complex and the second column is used to represent a member of the complex). Other PSICQUIC services will likely use a spoke model representation.

Please refer to our documentation above and documentation provided by other PSICQUIC providers if you plan to combine results from multiple services.

Main iRefIndex page


All iRefIndex Pages

Follow this link for a listing of all iRefIndex related pages (archived and current).