•http://www.hprd.org

Browse, query or BLAST? Or why not view one of the available protein-interaction networks? It's up to you. The Human Protein Reference Database (HPRD) is a user-friendly resource that lets you do it all. However you find your protein of choice, information on every possible feature — alternative names, function, sequence, domains, motifs, interactions, expression, localization, post-translational modifications, substrates and any disease association — is just a click away.

Couple all of this with links to external resources, such as the OMIM, Swiss-Prot, LocusLink and Unigene web sites, and this really is a unified protein bioinformatics platform.

Almost all of the information that is included in the HPRD has been obtained manually by biologists who read hundreds of thousands of publications and interpreted and analysed the data. Every protein is reviewed twice, but should you spot any errors, they can be reported online. The HPRD ontology should soon be fully compliant with that of the Gene Ontology consortium, and the HPRD data will eventually be downloadable.

This resource is a joint venture between Akhilesh Pandey's laboratory (http://pandeylab.bs.jhmi.edu) and the Institute of Bioinformatics (http://www.ibioinformatics.org), but contributions from 'the outside world' are also encouraged — so how about becoming a 'molecule authority' for your favourite protein? And if your protein of interest cannot be found, then let the curators know and they will annotate it for you.

With more than 3,000 proteins in the HPRD already, and that number expected to reach 10,000 by the end of 2003, this really is a great resource for anyone who is interested in the human proteome!