Approach to the evaluation of adverse antibiotic reactions in patients with cystic fibrosis.

Ann Allergy Asthma Immunol. 2016 Aug 30;

Authors: Petroni DH, Aitken ML, Ham E, Chung S, Menalia L, Altman MC, Ayars AG

Abstract
BACKGROUND: Adverse drug reactions (ADRs) to antibiotics in patients with cystic fibrosis (CF) are common and often mislabeled as allergies. The labeling of an antibiotic reaction as an allergy can lead to the use of antibiotics that are less efficacious, are more expensive, or have a greater risk of adverse effects.
OBJECTIVE: To establish a safe approach for the evaluation of ADRs to antibiotics in patients with CF to help clarify future use of these medications.
METHODS: Patients with CF whose antibiotic allergies were causing difficulty in their medical management were referred for an allergy evaluation that consisted of a thorough drug allergy history and antibiotic testing if appropriate. If the history was not consistent with a true hypersensitivity reaction (HSR) and test results were negative, the patient underwent a challenge to the offending agent(s) to rule out an HSR. Challenges were only performed if the medication was indicated for future use.
RESULTS: A total of 17 patients (mean age, 32.4 years) underwent a thorough allergy evaluation. A total of 17 antibiotic challenges were performed in 11 patients without a reaction consistent with an HSR or severe delayed reaction. Only 2 medications had a history consist with an HSR, and it was recommended that they undergo a desensitization procedure if the drug was required.
CONCLUSION: If treatment with appropriate antibiotics becomes difficult in patients with CF because of drug allergies, then referral to an allergist can help safely identify treatment options. Our findings suggest that a thorough evaluation by an allergy specialist can lead to more appropriate treatment options in patients with CF.

PMID: 27590641 [PubMed - as supplied by publisher]

HUMAN AIRWAY EPITHELIAL CELLS INVESTIGATED BY ATOMIC FORCE MICROSCOPY: A HINT TO CYSTIC FIBROSIS EPITHELIAL PATHOLOGY.

Exp Cell Res. 2016 Aug 30;

Authors: Lasalvia M, Castellani S, D'Antonio P, Perna G, Carbone A, Colia AL, Maffione AB, Capozzi V, Conese M

Abstract
The pathophysiology of cystic fibrosis (CF) airway disease stems from mutations in the CF Transmembrane Conductance Regulator (CFTR) gene, leading to a chronic respiratory disease. Actin cytoskeleton is disorganised in CF airway epithelial cells, likely contributing to the CF-associated basic defects, i.e.defective chloride secretion and sodium/fluid hypersorption. In this work, we aimed to find whether this alteration could be pointed out by means of Atomic Force Microscopy (AFM) investigation, as roughness and Young's elastic module. Moreover, we also sought to determine whether disorganisation of actin cytoskeleton is linked to hypersoption of apical fluid. Not only CFBE41o- (CFBE) cells, immortalised airway epithelial cells homozygous for the F508del CFTR allele showed a different morphology in comparison with 16HBE14o- (16HBE) epithelial cells, wild-type for CFTR, but also they displayed a lack of stress fibers, suggestive of a disorganised actin cytoskeleton. AFM measurements showed that CFBE cells presented a higher membrane roughness and decreased rigidity as compared with 16HBE cells. CFBE overexpressing wtCFTR became more elongated than the parental CFBE cell line and presented actin stress fibers. CFBE cells absorbed more fluid from the apical compartment. Study of fluid absorption with the F-actin-depolymerising agent Latrunculin B demonstrated that actin cytoskeletal disorganisation increased fluid absorption, an effect observed at higher magnitude in 16HBE than in CFBE cells. For the first time, we demonstrate that actin cytoskeleton disorganisation is reflected by AFM parameters in CF airway epithelial cells. Our data also strongly suggest that the lack of stress fibers is involved in at least one of the early step in CF pathophysiology at the levels of the airways, i.e. fluid hypersorption.

PMID: 27590528 [PubMed - as supplied by publisher]

FGF and EDA pathways control initiation and branching of distinct subsets of developing nasal glands.

Dev Biol. 2016 Aug 30;

Authors: May A, Headon D, Rice D, Noble A, Tucker AS

Abstract
Hypertrophy, hyperplasia and altered mucus secretion from the respiratory submucosal glands (SMG) are characteristics of airway diseases such as cystic fibrosis, asthma and chronic bronchitis. More commonly, hyper-secretion of the nasal SMGs contributes to allergic rhinitis and upper airway infection. Considering the role of these glands in disease states, there is a significant dearth in understanding the molecular signals that regulate SMG development and patterning. Due to the imperative role of FGF signalling during the development of other branched structures, we investigated the role of Fgf10 during initiation and branching morphogenesis of murine nasal SMGs. Fgf10 is expressed in the mesenchyme around developing SMGs while expression of its receptor Fgfr2 is seen within glandular epithelial cells. In the Fgf10 null embryo, Steno's gland and the maxillary sinus gland were completely absent while other neighbouring nasal glands showed normal duct elongation but defective branching. Interestingly, the medial nasal glands were present in Fgf10 homozygotes but missing in Fgfr2b mutants, with expression of Fgf7 specifically expressed around these developing glands, indicating that Fgf7 might compensate for loss of Fgf10 in this group of glands. Intriguingly the lateral nasal glands were only mildly affected by loss of FGF signalling, while these glands were missing in Eda mutant mice, where the Steno's and maxillary sinus gland developed as normal. This analysis reveals that regulation of nasal gland development is complex with different subsets of glands being regulated by different signalling pathways. This analysis helps shed light on the nasal gland defects observed in patients with hypohidrotic ectodermal dysplasia (HED) (defect EDA pathway) and LADD syndrome (defect FGFR2b pathway).

PMID: 27590203 [PubMed - as supplied by publisher]

Human kinetochores are swivel joints that mediate microtubule attachments.

Elife. 2016 Sep 3;5

Authors: Smith CA, McAinsh AD, Burroughs NJ

Abstract
Chromosome segregation is a mechanical process that requires assembly of the mitotic spindle - a dynamic microtubule-based force-generating machine. Connections to this spindle are mediated by sister kinetochore pairs, that form dynamic end-on attachments to microtubules emanating from opposite spindle poles. This bi-orientation generates forces that have been reported to stretch the kinetochore itself, which has been suggested to silence the spindle checkpoint and allow anaphase onset. We reveal using three dimensional tracking that the outer kinetochore domain can swivel around the inner kinetochore/centromere, which results in large reductions in intra-kinetochore distance (delta) when viewed in lower dimensions. We show that swivel provides a mechanical flexibility that enables kinetochores at the periphery of the spindle to engage microtubules. Swivel rather than delta reduces as cells approach anaphase, suggesting an organisational change linked to checkpoint satisfaction and/or obligatory changes in kinetochore mechanochemistry may occur before dissolution of sister chromatid cohesion.

PMID: 27591356 [PubMed - as supplied by publisher]

Predictive biophysical modeling and understanding of the dynamics of mRNA translation and its evolution.

Nucleic Acids Res. 2016 Sep 2;

Authors: Zur H, Tuller T

Abstract
mRNA translation is the fundamental process of decoding the information encoded in mRNA molecules by the ribosome for the synthesis of proteins. The centrality of this process in various biomedical disciplines such as cell biology, evolution and biotechnology, encouraged the development of dozens of mathematical and computational models of translation in recent years. These models aimed at capturing various biophysical aspects of the process. The objective of this review is to survey these models, focusing on those based and/or validated on real large-scale genomic data. We consider aspects such as the complexity of the models, the biophysical aspects they regard and the predictions they may provide. Furthermore, we survey the central systems biology discoveries reported on their basis. This review demonstrates the fundamental advantages of employing computational biophysical translation models in general, and discusses the relative advantages of the different approaches and the challenges in the field.

PMID: 27591251 [PubMed - as supplied by publisher]

Nonparametric dynamic modeling.

Math Biosci. 2016 Aug 30;

Authors: Faraji M, Voit EO

Abstract
Challenging as it typically is, the estimation of parameter values seems to be an unavoidable step in the design and implementation of any dynamic model. Here, we demonstrate that it is possible to set up, diagnose, and simulate dynamic models without the need to estimate parameter values, if the situation is favorable. Specifically, it is possible to establish nonparametric models for nonlinear compartment models, including metabolic pathway models, if sufficiently many high-quality time series data are available that describe the biological phenomenon under investigation in an appropriate and representative manner. The proposed nonparametric strategy is a variant of the method of Dynamic Flux Estimation (DFE), which permits the estimation of numerical flux profiles from metabolic time series data. However, instead of attempting to formulate these numerical profiles as explicit functions and to optimize their parameter values, as it is done in DFE, the metabolite and flux profiles are used here directly as a scaffold for a library from which values are interpolated and retrieved for the simulation of the differential equations describing the model. Beyond simulations, the proposed methods render it possible to determine steady states from non-steady state data, perform sensitivity analyses, and estimate the Jacobian of the system at a steady state.

PMID: 27590775 [PubMed - as supplied by publisher]

Plasmodium cynomolgi infections in rhesus macaques display clinical and parasitological features pertinent to modelling vivax malaria pathology and relapse infections.

Malar J. 2016;15(1):451

Authors: Joyner C, Moreno A, Meyer EV, Cabrera-Mora M, MaHPIC Consortium, Kissinger JC, Barnwell JW, Galinski MR

Abstract
BACKGROUND: Plasmodium vivax infections in humans or in new world monkeys pose research challenges that necessitate the use of alternative model systems. Plasmodium cynomolgi is a closely related species that shares genetic and biological characteristics with P. vivax, including relapses. Here, the haematological dynamics and clinical presentation of sporozoite-initiated P. cynomolgi infections in Macaca mulatta (rhesus macaques) are evaluated over a 100-day period.
METHODS: Five M. mulatta were inoculated with 2000 P. cynomolgi B strain sporozoites. Parasitological and haematological data were collected daily to study the clinical presentations of primary infections and relapses. Peripheral blood and bone marrow aspirates were collected at specific time points during infection for future and retrospective systems biology analyses.
RESULTS: Patent infections were observed between days 10 and 12, and the acute, primary infection consisted of parasitaemias ranging from 269,962 to 1,214,842 parasites/µl (4.42-19.5 % parasitaemia). All animals presented with anaemia, ranging from moderate (7-10 g/dl) to severe (<7 g/dl), based on peripheral haemoglobin concentrations. Minimum haemoglobin levels coincided with the clearance of parasites and peripheral reticulocytosis was evident at this time. Mild thrombocytopaenia (<150,000 platelets/µl) was observed in all animals, but unlike haemoglobin, platelets were lowest whenever peripheral parasitaemia peaked. The animals' conditions were classified as non-severe, severe or lethal (in one case) based upon their clinical presentation. The lethal phenotype presented uniquely with an exceptionally high parasitaemia (19.5 %) and lack of a modest reticulocyte release, which was observed in the other animals prior to acute manifestations. One or two relapses were observed in the four surviving animals, and these were characterized by significantly lower parasitaemias and minimal changes in clinical parameters compared to pre-infection values.
CONCLUSIONS: Rhesus macaque infections initiated by P. cynomolgi B strain sporozoites recapitulated pathology of human malaria, including anaemia and thrombocytopaenia, with inter-individual differences in disease severity. Importantly, this study provides an in-depth assessment of clinical and parasitological data, and shows that unlike the primary infections, the relapses did not cause clinical malaria. Notably, this body of research has provided experimental plans, large accessible datasets, and blood and bone marrow samples pertinent for ongoing and iterative systems biology investigations.

PMID: 27590312 [PubMed - as supplied by publisher]

BioCreative V BioC track overview: collaborative biocurator assistant task for BioGRID.

Database (Oxford). 2016;2016

Authors: Kim S, Islamaj Doğan R, Chatr-Aryamontri A, Chang CS, Oughtred R, Rust J, Batista-Navarro R, Carter J, Ananiadou S, Matos S, Santos A, Campos D, Oliveira JL, Singh O, Jonnagaddala J, Dai HJ, Su EC, Chang YC, Su YC, Chu CH, Chen CC, Hsu WL, Peng Y, Arighi C, Wu CH, Vijay-Shanker K, Aydın F, Hüsünbeyi ZM, Özgür A, Shin SY, Kwon D, Dolinski K, Tyers M, Wilbur WJ, Comeau DC

Abstract
BioC is a simple XML format for text, annotations and relations, and was developed to achieve interoperability for biomedical text processing. Following the success of BioC in BioCreative IV, the BioCreative V BioC track addressed a collaborative task to build an assistant system for BioGRID curation. In this paper, we describe the framework of the collaborative BioC task and discuss our findings based on the user survey. This track consisted of eight subtasks including gene/protein/organism named entity recognition, protein-protein/genetic interaction passage identification and annotation visualization. Using BioC as their data-sharing and communication medium, nine teams, world-wide, participated and contributed either new methods or improvements of existing tools to address different subtasks of the BioC track. Results from different teams were shared in BioC and made available to other teams as they addressed different subtasks of the track. In the end, all submitted runs were merged using a machine learning classifier to produce an optimized output. The biocurator assistant system was evaluated by four BioGRID curators in terms of practical usability. The curators' feedback was overall positive and highlighted the user-friendly design and the convenient gene/protein curation tool based on text mining.Database URL: http://www.biocreative.org/tasks/biocreative-v/track-1-bioc/.

PMID: 27589962 [PubMed - as supplied by publisher]

Overview of the interactive task in BioCreative V.

Database (Oxford). 2016;2016

Authors: Wang Q, S Abdul S, Almeida L, Ananiadou S, Balderas-Martínez YI, Batista-Navarro R, Campos D, Chilton L, Chou HJ, Contreras G, Cooper L, Dai HJ, Ferrell B, Fluck J, Gama-Castro S, George N, Gkoutos G, Irin AK, Jensen LJ, Jimenez S, Jue TR, Keseler I, Madan S, Matos S, McQuilton P, Milacic M, Mort M, Natarajan J, Pafilis E, Pereira E, Rao S, Rinaldi F, Rothfels K, Salgado D, Silva RM, Singh O, Stefancsik R, Su CH, Subramani S, Tadepally HD, Tsaprouni L, Vasilevsky N, Wang X, Chatr-Aryamontri A, Laulederkind SJ, Matis-Mitchell S, McEntyre J, Orchard S, Pundir S, Rodriguez-Esteban R, Van Auken K, Lu Z, Schaeffer M, Wu CH, Hirschman L, Arighi CN

Abstract
Fully automated text mining (TM) systems promote efficient literature searching, retrieval, and review but are not sufficient to produce ready-to-consume curated documents. These systems are not meant to replace biocurators, but instead to assist them in one or more literature curation steps. To do so, the user interface is an important aspect that needs to be considered for tool adoption. The BioCreative Interactive task (IAT) is a track designed for exploring user-system interactions, promoting development of useful TM tools, and providing a communication channel between the biocuration and the TM communities. In BioCreative V, the IAT track followed a format similar to previous interactive tracks, where the utility and usability of TM tools, as well as the generation of use cases, have been the focal points. The proposed curation tasks are user-centric and formally evaluated by biocurators. In BioCreative V IAT, seven TM systems and 43 biocurators participated. Two levels of user participation were offered to broaden curator involvement and obtain more feedback on usability aspects. The full level participation involved training on the system, curation of a set of documents with and without TM assistance, tracking of time-on-task, and completion of a user survey. The partial level participation was designed to focus on usability aspects of the interface and not the performance per se In this case, biocurators navigated the system by performing pre-designed tasks and then were asked whether they were able to achieve the task and the level of difficulty in completing the task. In this manuscript, we describe the development of the interactive task, from planning to execution and discuss major findings for the systems tested.Database URL: http://www.biocreative.org.

PMID: 27589961 [PubMed - as supplied by publisher]

Patients cured of acromegaly do not experience improvement of their skull deformities.

Pituitary. 2016 Sep 2;

Authors: Rick JW, Jahangiri A, Flanigan PM, Aghi MK

Abstract
PURPOSE: Acromegaly is a rare disease that is associated with many co-morbidities. This condition also causes progressive deformity of the skull which includes frontal bossing and cranial thickening. Surgical and/or medical management can cure this condition in many patients, but it is not understood if patients cured of acromegaly experience regression of their skull deformities.
METHODS: We performed a retrospective analysis on patients treated at our dedicated pituitary center from 2009 to 2014. We looked at all MRI images taken during the treatment of these patients and recorded measurements on eight skull dimensions. We then analyzed these measurements for changes over time.
RESULTS: 29 patients underwent curative treatment for acromegaly within our timeframe. The mean age for this population was 45.0 years old (range 19-70) and 55.2 % (n = 16) were female. All of these patients were treated with a transsphenoidal resection for a somatotropic pituitary adenoma. 9 (31.1%) of these patients required further medical therapy to be cured. We found statically significant variation in the coronal width of the sella turcica after therapy, which is likely attributable to changes from transsphenoidal surgery. None of the other dimensions had significant variation over time after cure.
CONCLUSION: Patients cured of acromegaly should not expect natural regression of their skull deformities. Our study suggests that both frontal bossing and cranial thickening do not return to normal after cure.

PMID: 27590786 [PubMed - as supplied by publisher]

Cutaneous leukocytoclastic vasculitis due to amoxicillin hypersensitivity.

Ann Allergy Asthma Immunol. 2016 Aug 30;

Authors: Sáenz de Santa María García M, Morales-Cabeza C, Noguerado-Mellado B, Rojas-Pérez-Ezquerra P, Zubeldia JM

PMID: 27590637 [PubMed - as supplied by publisher]

9 new pubmed citations were retrieved for your search. Click on the search hyperlink below to display the complete search results:

("orphan disease" OR "rare disease" OR "orphan diseases" OR "rare diseases")

These pubmed results were generated on 2016/09/03

PubMed comprises more than 24 million citations for biomedical literature from MEDLINE, life science journals, and online books. Citations may include links to full-text content from PubMed Central and publisher web sites.

7 new pubmed citations were retrieved for your search. Click on the search hyperlink below to display the complete search results:

"Cystic Fibrosis"

These pubmed results were generated on 2016/09/03

PubMed comprises more than 24 million citations for biomedical literature from MEDLINE, life science journals, and online books. Citations may include links to full-text content from PubMed Central and publisher web sites.

17 new pubmed citations were retrieved for your search. Click on the search hyperlink below to display the complete search results:

"Systems Biology"[Title/Abstract] AND ("2005/01/01"[PDAT] : "3000"[PDAT])

These pubmed results were generated on 2016/09/03

PubMed comprises more than 24 million citations for biomedical literature from MEDLINE, life science journals, and online books. Citations may include links to full-text content from PubMed Central and publisher web sites.

InteGO2: a web tool for measuring and visualizing gene semantic similarities using Gene Ontology.

BMC Genomics. 2016;17 Suppl 5:530

Authors: Peng J, Li H, Liu Y, Juan L, Jiang Q, Wang Y, Chen J

Abstract
BACKGROUND: The Gene Ontology (GO) has been used in high-throughput omics research as a major bioinformatics resource. The hierarchical structure of GO provides users a convenient platform for biological information abstraction and hypothesis testing. Computational methods have been developed to identify functionally similar genes. However, none of the existing measurements take into account all the rich information in GO. Similarly, using these existing methods, web-based applications have been constructed to compute gene functional similarities, and to provide pure text-based outputs. Without a graphical visualization interface, it is difficult for result interpretation.
RESULTS: We present InteGO2, a web tool that allows researchers to calculate the GO-based gene semantic similarities using seven widely used GO-based similarity measurements. Also, we provide an integrative measurement that synergistically integrates all the individual measurements to improve the overall performance. Using HTML5 and cytoscape.js, we provide a graphical interface in InteGO2 to visualize the resulting gene functional association networks.
CONCLUSIONS: InteGO2 is an easy-to-use HTML5 based web tool. With it, researchers can measure gene or gene product functional similarity conveniently, and visualize the network of functional interactions in a graphical interface. InteGO2 can be accessed via http://mlg.hit.edu.cn:8089/ .

PMID: 27586009 [PubMed - in process]

Application of pharmacologically induced transcriptomic profiles to interrogate PI3K-Akt-mTOR pathway activity associated with cancer patient prognosis.

Oncotarget. 2016 Aug 31;

Authors: Ung MH, Wang GL, Varn FS, Cheng C

Abstract
The PI3K-Akt-mTOR signaling pathway has been identified as a key driver of carcinogenesis in several cancer types. As such, a major area of focus in cancer biology is the development of genomic biomarkers that can measure the activity level of the PI3K-Akt-mTOR pathway. In this study, we systematically estimate PI3K-Akt-mTOR pathway activity in breast primary tumor samples using transcriptomic profiles derived from drug treatment in MCF7 cell lines. We demonstrate that gene expression profiles derived from chemically-induced protein inhibition allows us to measure PI3K-Akt-mTOR pathway activity in patient tumor samples. With this approach, we predict prognosis and response to chemotherapy in cancer patients, and screen for potential pharmacological modulators of PI3K-Akt-mTOR pathway inhibitors.

PMID: 27589846 [PubMed - as supplied by publisher]

Next-generation sequencing in neuromuscular diseases.

Curr Opin Neurol. 2016 Oct;29(5):527-536

Authors: Efthymiou S, Manole A, Houlden H

Abstract
PURPOSE OF REVIEW: Neuromuscular diseases are clinically and genetically heterogeneous and probably contain the greatest proportion of causative Mendelian defects than any other group of conditions. These disorders affect muscle and/or nerves with neonatal, childhood or adulthood onset, with significant disability and early mortality. Along with heterogeneity, unidentified and often very large genes require complementary and comprehensive methods in routine molecular diagnosis. Inevitably, this leads to increased diagnostic delays and challenges in the interpretation of genetic variants.
RECENT FINDINGS: The application of next-generation sequencing, as a research and diagnostic strategy, has made significant progress into solving many of these problems. The analysis of these data is by no means simple, and the clinical input is essential to interpret results.
SUMMARY: In this review, we describe using examples the recent advances in the genetic diagnosis of neuromuscular disorders, in research and clinical practice and the latest developments that are underway in next-generation sequencing. We also discuss the latest collaborative initiatives such as the Genomics England (Department of Health, UK) genome sequencing project that combine rare disease clinical phenotyping with genomics, with the aim of defining the vast majority of rare disease genes in patients as well as modifying risks and pharmacogenomics factors.

PMID: 27588584 [PubMed - as supplied by publisher]

Advancing Systems Biology in the International Conference on Intelligent Biology and Medicine (ICIBM) 2015.

BMC Syst Biol. 2016;10 Suppl 3:61

Authors: Zhao Z, Liu Y, Huang Y, Huang K, Ruan J

Abstract
The 2015 International Conference on Intelligent Biology and Medicine (ICIBM 2015) was held on November 13-15, 2015 in Indianapolis, Indiana, USA. ICIBM 2015 included eight scientific sessions, three tutorial sessions, one poster session, and four keynote presentations that covered the frontier research in broad areas related to bioinformatics, systems biology, big data science, biomedical informatics, pharmacogenomics, and intelligent computing. Here, we present a summary of the 10 research articles that were selected from ICIBM 2015 and included in the supplement to BMC Systems Biology.

PMID: 27587087 [PubMed - in process]

Use of large-scale veterinary data for the investigation of antimicrobial prescribing practices in equine medicine.

Equine Vet J. 2016 Sep 2;

Authors: Welsh CE, Parkin TD, Marshall JF

Abstract
BACKGROUND: As antimicrobial resistant bacterial strains continue to emerge and spread in human and animal populations, understanding prescription practices is key in benchmarking current performance and setting goals. Antimicrobial prescription in companion veterinary species is widespread, but is neither monitored nor restricted in the US and Canada. The veterinary use of certain antimicrobial classes is discouraged in some countries, in the hope of preserving efficacy for serious human infections.
OBJECTIVES: The aim of this study was to ascertain the rate of prescription of a number of 'reserved' antimicrobials in a first-opinion US and Canadian horse cohort, and identify trends in their empirical use.
STUDY DESIGN: Retrospective cohort study.
METHODS: A large convenience sample of electronic medical records (2006 to 2012) were interrogated using text mining to identify enrofloxacin, clarithromycin and ceftiofur prescriptions. Time series analysis and logistic regression were used to identify trends and risk factors for prescription.
RESULTS: Prescription of these antimicrobials as a first-line intervention, without culture and sensitivity testing, was common in this population. Enrofloxacin prescriptions were found to increase over the study period, and there was evidence of either a reducing, or static trend in the proportion of reserved antimicrobial prescriptions informed by culture and sensitivity testing.
MAIN LIMITATIONS: Dose adequacy could not be included due to the nature of the data used.
CONCLUSIONS: Empirical use of reserved antimicrobials was common in this population, and further advice and guidance should be issued to first-opinion veterinarians to safeguard antimicrobial efficacy. This article is protected by copyright. All rights reserved.

PMID: 27589226 [PubMed - as supplied by publisher]

Leveraging syntactic and semantic graph kernels to extract pharmacokinetic drug drug interactions from biomedical literature.

BMC Syst Biol. 2016;10 Suppl 3:67

Authors: Zhang Y, Wu HY, Xu J, Wang J, Soysal E, Li L, Xu H

Abstract
BACKGROUND: Information about drug-drug interactions (DDIs) supported by scientific evidence is crucial for establishing computational knowledge bases for applications like pharmacovigilance. Since new reports of DDIs are rapidly accumulating in the scientific literature, text-mining techniques for automatic DDI extraction are critical. We propose a novel approach for automated pharmacokinetic (PK) DDI detection that incorporates syntactic and semantic information into graph kernels, to address the problem of sparseness associated with syntactic-structural approaches. First, we used a novel all-path graph kernel using shallow semantic representation of sentences. Next, we statistically integrated fine-granular semantic classes into the dependency and shallow semantic graphs.
RESULTS: When evaluated on the PK DDI corpus, our approach significantly outperformed the original all-path graph kernel that is based on dependency structure. Our system that combined dependency graph kernel with semantic classes achieved the best F-scores of 81.94 % for in vivo PK DDIs and 69.34 % for in vitro PK DDIs, respectively. Further, combining shallow semantic graph kernel with semantic classes achieved the highest precisions of 84.88 % for in vivo PK DDIs and 74.83 % for in vitro PK DDIs, respectively.
CONCLUSIONS: We presented a graph kernel based approach to combine syntactic and semantic information for extracting pharmacokinetic DDIs from Biomedical Literature. Experimental results showed that our proposed approach could extract PK DDIs from literature effectively, which significantly enhanced the performance of the original all-path graph kernel based on dependency structure.

PMID: 27585838 [PubMed - in process]