PubMed Health⌕ Search

Biomedical subjects

Dietrich Rebholz-Schuhmann

Publications and source records attributed to Dietrich Rebholz-Schuhmann.

9 recordsLinked to original sources
PubMed2006

GOAnnotator: linking protein GO annotations to evidence text.

BACKGROUND: Annotation of proteins with gene ontology (GO) terms is ongoing work and a complex task. Manual GO annotation is precise and precious, but it is time-consuming. Therefore, instead of curated annotations most of the proteins come with uncurated annotations, which have been generated automatically. Text-mining systems that use literature for automatic annotation have been proposed but they do not satisfy the high quality expectations of curators. RESULTS: In this paper we describe an approach that links uncurated annotations to text extracted from literature. The selection of the text is based on the similarity of the text to the term from the uncurated annotation. Besides substantiating the uncurated annotations, the extracted texts also lead to novel annotations. In addition, the approach uses the GO hierarchy to achieve high precision. Our approach is integrated into GOAnnotator, a tool that assists the curation process for GO annotation of UniProt proteins. CONCLUSION: The GO curators assessed GOAnnotator with a set of 66 distinct UniProt/SwissProt proteins with uncurated annotations. GOAnnotator provided correct evidence text at 93% precision. This high precision results from using the GO hierarchy to only select GO terms similar to GO terms from uncurated annotations in GOA. Our approach is the first one to achieve high precision, which is crucial for the efficient support of GO curators. GOAnnotator was implemented as a web tool that is freely available at http://xldb.di.fc.ul.pt/rebil/tools/goa/.

Journal Article↗
PubMed2006

Dealing with repetitions in sequencing by hybridization.

DNA sequencing by hybridization (SBH) induces errors in the biochemical experiment. Some of them are random and disappear when the experiment is repeated. Others are systematic, involving repetitions in the probes of the target sequence. A good method for solving SBH problems must deal with both types of errors. In this work we propose a new hybrid genetic algorithm for isothermic and standard sequencing that incorporates the concept of structured combinations. The algorithm is then compared with other methods designed for handling errors that arise in standard and isothermic SBH approaches. DNA sequences used for testing are taken from GenBank. The set of instances for testing was divided into two groups. The first group consisted of sequences containing positive and negative errors in the spectrum, at a rate of up to 20%, excluding errors coming from repetitions. The second group consisted of sequences containing repeated oligonucleotides, and containing additional errors up to 5% added into the spectra. Our new method outperforms the best alternative procedures for both data sets. Moreover, the method produces solutions exhibiting extremely high degree of similarity to the target sequences in the cases without repetitions, which is an important outcome for biologists. The spectra prepared from the sequences taken from GenBank are available on our website http://bio.cs.put.poznan.pl/.

Algorithms↗
PubMed2006

Using argumentation to extract key sentences from biomedical abstracts.

PROBLEM: key word assignment has been largely used in MEDLINE to provide an indicative "gist" of the content of articles and to help retrieving biomedical articles. Abstracts are also used for this purpose. However with usually more than 300 words, MEDLINE abstracts can still be regarded as long documents; therefore we design a system to select a unique key sentence. This key sentence must be indicative of the article's content and we assume that abstract's conclusions are good candidates. We design and assess the performance of an automatic key sentence selector, which classifies sentences into four argumentative moves: PURPOSE, METHODS, RESULTS and CONCLUSION METHODS: we rely on Bayesian classifiers trained on automatically acquired data. Features representation, selection and weighting are reported and classification effectiveness is evaluated on the four classes using confusion matrices. We also explore the use of simple heuristics to take the position of sentences into account. Recall, precision and F-scores are computed for the CONCLUSION class. For the CONCLUSION class, the F-score reaches 84%. Automatic argumentative classification using Bayesian learners is feasible on MEDLINE abstracts and should help user navigation in such repositories.

Abstracting and Indexing↗
PubMed2005

Distributed modules for text annotation and IE applied to the biomedical domain.

Biological databases contain facts from scientific literature that have been curated by hand to ensure high quality. Curation is time-consuming and can be supported by information extraction methods. We present a server software infrastructure which allows to easily plug in modules to identify biologically interesting pieces of text to be then presented in a web interface to the curator. There are modules which identify UniProt, UMLS and GO terminology, gene and protein names, mutations and protein-protein interactions. UniProt, UMLS and GO concepts are automatically linked to the original source. The module for mutations is based on syntax patterns and the one for protein-protein interactions relies on chunk parsing. All modules work as separate servers possibly distributed on different machines and can be combined into processing pipelines as necessary. Communication is based on XML annotated text streams, each server processing the XML elements it is designed for, and possibly adding more information in the form of XML annotation. The server and the underlying software are available to the public.

Abstracting and Indexing↗
PubMed2005

Extracting key sentences with latent argumentative structuring.

PROBLEM: Key word assignment has been largely used in MEDLINE to provide an indicative "gist" of the content of articles. Abstracts are also used for this purpose. However with usually more than 300 words, abstracts can still be regarded as long documents; therefore we design a system to select a unique key sentence. This key sentence must be indicative of the article's content and we assume that abstract's conclusions are good candidates. We design and assess the performance of an automatic key sentence selector, which classifies sentences into 4 argumentative moves: PURPOSE, METHODS, RESULTS and CONCLUSION. METHODS: We rely on Bayesian classifiers trained on automatically acquired data. Features representation, selection and weighting are reported and classification effectiveness is evaluated on the four classes using confusion matrices. We also explore the use of simple heuristics to take the position of sentences into account. Recall, precision and F-scores are computed for the CONCLUSION class. For the CONCLUSION class, the F-score reaches 84%. Automatic argumentative classification is feasible on MEDLINE abstracts and should help user navigation in such repositories.

Bayes Theorem↗
PubMed2004

Automatic extraction of mutations from Medline and cross-validation with OMIM.

Mutations help us to understand the molecular origins of diseases. Researchers, therefore, both publish and seek disease-relevant mutations in public databases and in scientific literature, e.g. Medline. The retrieval tends to be time-consuming and incomplete. Automated screening of the literature is more efficient. We developed extraction methods (called MEMA) that scan Medline abstracts for mutations. MEMA identified 24,351 singleton mutations in conjunction with a HUGO gene name out of 16,728 abstracts. From a sample of 100 abstracts we estimated the recall for the identification of mutation-gene pairs to 35% at a precision of 93%. Recall for the mutation detection alone was >67% with a precision rate of >96%. This shows that our system produces reliable data. The subset consisting of protein sequence mutations (PSMs) from MEMA was compared to the entries in OMIM (20,503 entries versus 6699, respectively). We found 1826 PSM-gene pairs to be in common to both datasets (cross-validated). This is 27% of all PSM-gene pairs in OMIM and 91% of those pairs from OMIM which co-occur in at least one Medline abstract. We conclude that Medline covers a large portion of the mutations known to OMIM. Another large portion could be artificially produced mutations from mutagenesis experiments. Access to the database of extracted mutation-gene pairs is available through the web pages of the EBI (refer to http://www.ebi. ac.uk/rebholz/index.html).

Animals↗
PubMed2003

Computer-assisted generation of a protein-interaction database for nuclear receptors.

With the increasing amount of biological data available, automated methods for information retrieval become necessary. We employed computer-assisted text mining to retrieve all protein-protein interactions for nuclear receptors from MEDLINE in a systematic way. A dictionary of protein names and of terms denoting interactions was generated, and trioccurrences of two protein names and one interaction term in one sentence were retrieved. Abstracts containing at least one such trioccurrence were manually checked by biologists to select the relevant interactions out of the automatically extracted data. In total, 4360 abstracts were retrieved containing data on protein interactions for nuclear receptors. The resulting database contains all reported protein interactions involving nuclear receptors from 1966 to September 2001. Remarkably, the annual increase in number of reported interactors for nuclear receptors has been following an exponential growth curve in the years 1991 to 2001. Apparent in the data set is the high complexity of protein interactions for nuclear receptors. The number of interactions correlates with the number of published papers for a given receptor, suggesting that the number of reported interactors is a reflection of the intensity of research dedicated to a given receptor. Indeed, comparison of the retrieved data to a systematic yeast two-hybrid-based interaction analysis suggests that most NRs are similar with respect to the number of interacting proteins. The data set obtained serves as a source for information on NR interactions, as well as a reference data set for the improvement of advanced text-mining methods.

Computers↗