Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: LocText: relation extraction of protein localizations to assist database curation

Fig. 3

LocText pipeline. The input are text documents (e.g. PubMed). First, the STRING Tagger recognizes named entities (NER): proteins (green in the example; linked to UniProtKB), cellular localizations (pink; linked to GO), and organisms (yellow; linked to NCBI Taxonomy). Then, the relation extractor (RE) of LocText resolves which proteins and localizations are related (as in “localized in”). The output is a list of text-mined relationships (GO annotations) linked to the original text sources (e.g. PMIDs)

Back to article page