Computer-assisted curation of a human regulatory core network from the biological literature.

MOTIVATION: A highly interlinked network of transcription factors (TFs) orchestrates the context-dependent expression of human genes. ChIP-chip experiments that interrogate the binding of particular TFs to genomic regions are used to reconstruct gene regulatory networks at genome-scale, but are plagued by high false-positive rates. Meanwhile, a large body of knowledge on high-quality regulatory interactions remains largely unexplored, as it is available only in natural language descriptions scattered over millions of scientific publications. Such data are hard to extract and regulatory data currently contain together only 503 regulatory relations between human TFs. RESULTS: We developed a text-mining-assisted workflow to systematically extract knowledge about regulatory interactions between human TFs from the biological literature. We applied this workflow to the entire Medline, which helped us to identify more than 45 000 sentences potentially describing such relationships. We ranked these sentences by a machine-learning approach. The top-2500 sentences contained ∼900 sentences that encompass relations already known in databases. By manually curating the remaining 1625 top-ranking sentences, we obtained more than 300 validated regulatory relationships that were not present in a regulatory database before. Full-text curation allowed us to obtain detailed information on the strength of experimental evidences supporting a relationship. CONCLUSIONS: We were able to increase curated information about the human core transcriptional network by >60% compared with the current content of regulatory databases. We observed improved performance when using the network for disease gene prioritization compared with the state-of-the-art. AVAILABILITY AND IMPLEMENTATION: Web-service is freely accessible at http://fastforward.sys-bio.net/. CONTACT: leser@informatik.hu-berlin.de or nils.bluethgen@charite.de SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Genome comparison of the epiphytic bacteria Erwinia billingiae and E. tasmaniensis with the pear pathogen E. pyrifoliae.

BACKGROUND: The genus Erwinia includes plant-associated pathogenic and non-pathogenic Enterobacteria. Important pathogens such as Erwinia amylovora, the causative agent of fire blight and E. pyrifoliae causing bacterial shoot blight of pear in Asia belong to this genus. The species E. tasmaniensis and E. billingiae are epiphytic bacteria and may represent antagonists for biocontrol of fire blight. The presence of genes that are putatively involved in virulence in E. amylovora and E. pyrifoliae is of special interest for these species in consequence. RESULTS: Here we provide the complete genome sequences of the pathogenic E. pyrifoliae strain Ep1/96 with a size of 4.1 Mb and of the non-pathogenic species E. billingiae strain Eb661 with a size of 5.4 Mb, de novo determined by conventional Sanger sequencing and next generation sequencing techniques. Genome comparison reveals large inversions resulting from homologous recombination events. Furthermore, comparison of deduced proteins highlights a relation of E. billingiae strain Eb661 to E. tasmaniensis strain Et1/99 and a distance to E. pyrifoliae for the overall gene content as well as for the presence of encoded proteins representing virulence factors for the pathogenic species. Pathogenicity of E. pyrifoliae is supposed to have evolved by accumulation of potential virulence factors. E. pyrifoliae carries factors for type III secretion and cell invasion. Other genes described as virulence factors for E. amylovora are involved in the production of exopolysaccharides, the utilization of plant metabolites such as sorbitol and sucrose. Some virulence-associated genes of the pathogenic species are present in E. tasmaniensis but mostly absent in E. billingiae. CONCLUSION: The data of the genome analyses correspond to the pathogenic lifestyle of E. pyrifoliae and underlines the epiphytic localization of E. tasmaniensis and E. billingiae as a saprophyte.

Identification of whole pathogenic cells by monoclonal antibodies generated against a specific peptide from an immunogenic cell wall protein.

We selected the immunogenic cell wall ß-(1,3)-glucosyltransferase Bgl2p from Candida albicans as a target protein for the production of antibodies. We identified a unique peptide sequence in the protein and generated monoclonal anti- C. albicans Bgl2p antibodies, which bound in particular to whole C. albicans cells.

Ovarian stimulation affects the levels of regulatory endometrial NK cells and angiogenic cytokine VEGF.

PROBLEM: Endometrial NK cells play a critical role in uterine vascularization producing angiogenic factors. Impact of ovarian stimulation on endometrial expression of NK cells and VEGF in normal fertile oocyte donors and the effect of endometrial injury treatment on these parameters have been investigated. METHOD OF STUDY: Endometrial tissue was obtained from oocyte donors during natural and stimulated cycles. NK cell subsets were measured by flow cytometry. VEGF was determined by ELISA and flow cytometry. Endometrial angiogenic parameters were determined by ultrasound Doppler. Local injury was induced by scratching endometrial tissue previous to implantation window. RESULTS: Ovarian stimulation decreased endometrial levels of NK cells and vascularization index but increased VEGF levels. Local injury normalized only the CD56(+) NK cell count. CONCLUSION: Hormonal therapy for ovarian stimulation may be associated with poor endometrial vascularization. Local injury before the implantation window seems not to influence endometrial angiogenic parameters altered by ovarian stimulation.

Prioritising the prevention of medication handling errors.

OBJECTIVE: Medication errors are frequent in a hospital setting and often caused by inappropriate drug handling. Systematic strategies for their prevention however are still lacking. We developed and applied a classification model to categorise medication handling errors and defined the urgency of correction on the basis of these findings. SETTING: Nurses on medical wards (including intensive and intermediate care units) of a 1,680-bed teaching hospital. METHOD: In a prospective observational study we evaluated the prevalence of 20 predefined medication handling errors on the ward. In a concurrent questionnaire survey, we assessed the knowledge of the nurses on medication handling. The severity of errors observed in individual areas was scored considering prevalence, potential risk of an error, and the involved drug. These scores and the prevalence of corresponding knowledge deficits were used to define the urgency of preventive strategies according to a four-field decision matrix. MAIN OUTCOME MEASURE: Prevalence and potential risk of medication handling errors, corresponding knowledge deficits in nurses committing the errors, and priority of quality improvement. RESULTS: In 1,376 observed processes 833 medication handling errors were detected. Errors concerning preparation (mean 0.88 errors per observed process [95% CI: 0.81-0.96], N = 645) were more frequent than administration errors (0.36 [0.32-0.41], N = 701, P < 0.001). Parenteral drugs (1.10 [1.00-1.19], N = 492) were more often involved in errors than enteral drugs (0.32 [0.28-0.36], N = 794, P < 0.001). Of the 833 observed medication errors 30.9% concerned processes of high risk, 19.0% of moderate risk, and 50.1% of low risk. Of these errors 11.4% were caused by critical dose drugs, 81.6% by uncomplicated drugs, and 6.9% by nutritional supplements or diluents without active ingredient. According to the decision matrix that also considered knowledge deficits two error types concerning enteral drugs (flaws in light protection and prescribing information) were given maximum priority for quality improvement. For parenteral drugs five errors (incompatibilities, flaws in hygiene, duration of administration, check for visible abnormalities, and again prescribing information) appeared most important. CONCLUSION: We successfully applied a newly developed classification model to prioritise medication handling errors for prevention strategies.

Prevention of intravenous drug incompatibilities in an intensive care unit.

PURPOSE: The frequency of drug administration errors and incompatibilities between intravenous drugs before and after an intervention in an intensive care unit (ICU) is discussed. METHODS: Critically ill adult patients with intoxications, multiorgan failure, and serious infections were included in a retrospective analysis and in a prospective two-period, one-sequence study. In the retrospective analysis, the most frequent brands of i.v. medications used in the ICU of a gastroenterologic department in a teaching hospital were identified. All possible combinations and resulting incompatibilities were defined. Based on the results, a standard operating procedure (SOP) was established to prevent frequent and well-documented incompatibilities among i.v. medications. In the prospective study, trained pharmacy students assessed incompatible coinfusions before and after SOP implementation. RESULTS: In the retrospective analysis of 100 patients, 3617 brands of drug pairs were potentially given concurrently through one i.v. line and 7.2% of the drug pairs were incompatible. Antibiotics, such as piperacillin-tazobactam and imipenem-cilastatin, were the most frequent incompatible drug pairs. The newly developed SOP mandated that administration of these drugs be separated from all other drugs and suggested the use of an idle i.v. line for infusion whenever possible. In the prospective study of 50 patients, the frequency of incompatible drug pairs was reduced by the time of intervention from 5.8% to 2.4%. Incompatible drug pairs that were governed by the new SOP were reduced from 1.9% to 0.5%. CONCLUSION: Administration of incompatible i.v. drugs in critically ill patients was frequent but significantly reduced by procedural interventions with SOPs.