Extracting the names of genes and gene products with a hidden Markov model

Abstract

\~e report the results of a study into the use of a linear interpolating hidden Marker model (HMM) for the task of extra.('ting lxw]mi(:al |;erminology fl:om MEDLINE al)stra('ts and texl;s in the molecular-bioh)gy domain. Tiffs is the first stage isl a. system that will exl;ra('l; evenl; information for automatically ut)da.ting 1)ioh)gy databases. We trained the HMM entirely with 1)igrams based (m lexical and character features in a relatively small corpus of 100 MED-LINE abstract;s that were ma.rked-ul) l)y (lomain experts wil;h term (:lasses su(:h as t)rol;eins and DNA. I.Jsing cross-validation methods we a(:]fieved a,n ].e-score of 0.73 and we (',xmnine the ('ontrilmtion made by each 1)art of the interl)olation model to overconfing (la.ta Sl)arsen('.ss.

DEEPER: A Full Parsing Based Approach to Protein Relation Extraction

“…The rst step is the recognition of the protein names themselves (see e.g. [3,6,15]). As the focus of this paper is on the mining of interactions, we assume that protein name recognition has already taken place.…”

“…More and more relevant information is becoming available on the web, in particular in literature databases such as MEDLINE 3 , in ontological resources such as the Gene Ontology 4 , and in specialized structured databases such as IntAct 5 . The unstructured information in scientic publications poses the biggest challenge to biologists who are interested in specic gene or protein interactions, as they are forced to spend a tremendous amount of time reviewing articles looking for the information they need.…”

Discovering semantic biomedical relations utilizing the Web

“…In Fukuda et al [1998], protein names are identified in biological papers using hand-coded rules. On the other hand, in Collier et al [2000], supervised learning methods based on Hidden Markov Models are used. Subramaniam et al [2003] have developed the BioAnnotator system, which is part of the current Relation Extraction system, and uses rules and dictionary lookup for identifying and classifying biological terms.…”

Getting to the (c)ore of knowledge: mining biomedical literature

“…Hidden Markov Models (HMMs) [64] can learn a lexicon and context as well by computing the probability that a sequence of specific words surround or constitute a molecule name. The expert just has to identify examples, while the HMM learns the patterns to apply to new sequences of words.…”