top of page


BioCodeKb - Bioinformatics Knowledgebase

IPPRED is a web based server to conclude protein-protein interactions through homology search between candidate proteins and those described as interacting. This simple result allows to propose or to validate potential interactions.

The knowledge of interactions between proteins is essential for understanding cell functions. The identification of these interactions is the subject of many projects ranging from small to large scale. These projects use several experimental approaches such as two hybrid screening or purification of complexes followed by the characterization of their components. In order to help biologists we have developed IPPRED, a web based server.

In order to infer interactions among a set of submitted candidate proteins, IPPRED first identifies homology relationships between those proteins and proteins that are involved in experimentally characterized interactions. These recommendation interactions are extracted from the literature and from public databases like BIND. To enable fast and reliable inference, IPPRED relies on a local database which have a comprehensive and non-redundant set of described interactions and the corresponding sequences. BLAST program is used for similarity searches and succeding prediction of homology. The IPPRED output is a list of inferred interactions between candidate proteins. For each inferred pair, the server provides information about the reference interactions which were used for the inference, BLAST similarity scores and the location of putative homologous domains along the sequences.

Query submission is done through a web form which accepts one or more protein sequences in different formats and allows the user to specify a threshold for BLAST similarity to be used for homology prediction. Upon availability of the results, an email with a link to the results page is sent to the user. Those results are given with two levels of details;

The first one is the main results page which lists all the inferred interactions, for each of them it displays the two partners, the number of described interactions used for this inference and a link to a detailed page

The second one is the detailed page. This page is available for each inferred interaction and contains a graphical view which shows the location of BLAST matches for all the reference interactions

A colour schema is used to reflect the level of similarity. For each reference interaction, links are provided which give access to native BLAST output, description of the proteins and the source of the interaction.

Results are kept available for 2 weeks on the server and then deleted. It is possible to save inferred interactions in text or HTML format. IPPRED offers an new service which is complementary to the resources made available by the providers of protein interactions databases. Designed for the biologists, IPPRED is used to identify new potential interactions among a set of proteins. It also helps for the validation and annotation of experimental results on protein– protein interactions. In further versions, IPPRED will handle information about domain composition of proteins and orthology relationships in future plans.


Need to learn more about IPPRED and much more?

To learn Bioinformatics, analysis, tools, biological databases, Computational Biology, Bioinformatics Programming in Python & R through interactive video courses and tutorials, Join BioCode.

bottom of page