Skip to main content

Automatic extraction of keywords for biological sequences

Objective



The increasing rate at which biological sequences arrive at public databases makes difficult high quality manual annotation. We propose an automatic method for finding keywords to descrive novel sequences; the method is based on the distribution of words in MEDLINE abstracts related to the sequence. The method will be used by database annotators at the European Bioinformatics Institute (Cambridge, UK). We will also provide a World Wide Web (WWW) server for European molecular biologists which gives 'intelligent' links between sequence and related MEDLINE articles.

Funding Scheme

RGI - Research grants (individual fellowships)

Coordinator

EUROPEAN MOLECULAR BIOLOGY LABORATORY
Address
Wellcome Trust Genome Campus, Hinxton Hall
CB10 1SD Saffron Walden
United Kingdom

Participants (1)

Not available
Spain