Rule-Based Protein Term Identification with Help from Automatic Species Tagging

Abstract

Abstract. In biomedical articles, terms often refer to different protein entities. For example, an arbitrary occurrence of term p53 might denote thousands of proteins across a number of species. A human annotator is able to resolve this ambiguity relatively easily, by looking at its context and if necessary, by searching an appropriate protein database. However, this phenomenon may cause much trouble to a text mining system, which does not understand human languages and hence can not identify the correct protein that the term refers to. In this paper, we present a Term Identification system which automatically assigns unique identifiers, as found in a protein database, to ambiguous protein mentions in texts. Unlike other solutions described in literature, which only work on gene/protein mentions on a specific model organism, our system is able to tackle protein mentions across many species, by integrating a machine-learning based species tagger. We have compared the performance of our automatic system to that of human annotators, with very promising results.

Species disambiguation for biomedical term identification

“…As noted in previous work (Krauthammer and Nenadic, 2004;Chen et al, 2005;Wang, 2007), determining the correct species for the protein mentions is a very important step towards TI. However, as far as we know, there has been little work in species disambiguation and in to what extent resolving species ambiguity can help TI.…”

Comparing Usability of Matching Techniques for Normalising Biomedical Named Entities

“…One direction for future work is to conduct more curation experiments so that the variability between curators can be smoothed (e.g., some curators may prefer seeing more accurate NLP output whereas others may prefer higher recall). Meanwhile, we plan to improve the matching systems by integrating ontology processors and species disambiguators [7].…”

“…j Tissues were to be assigned to MeSH k IDs and proteins to RefSeq IDs. We selected only human proteins for this experiment, because although species is a major source of ambiguity in biological entities [7], we wanted to focus on investigating how matching techniques affect curation speed in this work. Curation was carried out using an in-house curation tool (as shown in Figure 1).…”

“…A typical TN system consists of three components: an ontology processor, which expands or prunes the reference ontology; a string matcher, which compares entity mentions in articles against entries in the processed ontology; and finally a filter (or a disambiguator) that removes false positive identifiers using rules or statistical models [6,7]. The string matcher is arguably the core component: a matcher that searches a database and retrieves entries that exactly match an entity mention can form a simple TN system.…”

Assisted Curation: Does Text Mining Really Help?

“…Species are assigned to proteins using a machine learning based tagger trained on contextual and species word features [14]. The species information and a set of heuristics are used to choose the most probable identifiers from the set of candidates proposed by the matcher.…”