A hybrid model for named entity recognition using unstructured medical text

Named entity recognition (NER) is an essential step in the process of information extraction within text mining. This paper proposes a technique to extract drug named entities from unstructured and informal medical text using a hybrid model of lexicon-based and rule-based techniques. In the proposed model, a lexicon is first used as the initial step to detect drug named entities. Inference rules are then deployed to further extract undetected drug names. The designed rules employ part of speech tags and morphological features for drug name detection. The proposed hybrid model is evaluated using a benchmark data set from the i2b2 2009 medication challenge, and is able to achieve an f-score of 66.97%.

Enhancing medical named entity recognition with an extended segment representation technique

Objective : The objective of this paper is to formulate an extended segment representation (SR) technique to enhance named entity recognition (NER) in medical applications.

Methods : An extension to the IOBES (Inside/Outside/Begin/End/Single) SR technique is formulated. In the proposed extension, a new class is assigned to words that do not belong to a named entity (NE) in one context but appear as an NE in other contexts. Ambiguity in such cases can negatively affect the results of classification-based NER techniques. Assigning a separate class to words that can potentially cause ambiguity in NER allows a classifier to detect NEs more accurately; therefore increasing classification accuracy.

Results : The proposed SR technique is evaluated using the i2b2 2010 medical challenge data set with eight different classifiers. Each classifier is trained separately to extract three different medical NEs, namely treatment, problem, and test. From the three experimental results, the extended SR technique is able to improve the average F1-measure results pertaining to seven out of eight classifiers. The kNN classifier shows an average reduction of 0.18% across three experiments, while the C4.5 classifier records an average improvement of 9.33%.

Classification ensemble to improve medical named entity recognition

An accurate Named Entity Recognition (NER) is important for knowledge discovery in text mining. This paper proposes an ensemble machine learning approach to recognise Named Entities (NEs) from unstructured and informal medical text. Specifically, Conditional Random Field (CRF) and Maximum Entropy (ME) classifiers are applied individually to the test data set from the i2b2 2010 medication challenge. Each classifier is trained using a different set of features. The first set focuses on the contextual features of the data, while the second concentrates on the linguistic features of each word. The results of the two classifiers are then combined. The proposed approach achieves an f-score of 81.8%, showing a considerable improvement over the results from CRF and ME classifiers individually which achieve f-scores of 76% and 66.3% for the same data set, respectively.

Enhancement of medical named entity recognition using graph-based features

Named Entity Recognition (NER) is a crucial step in text mining. This paper proposes a new graph-based technique for representing unstructured medical text. The new representation is used to extract discriminative features that are able to enhance the NER performance. To evaluate the usefulness of the proposed graph-based technique, the i2b2 medication challenge data set is used. Specifically, the 'treatment' named entities are extracted for evaluation using six different classifiers. The F-measure results of five classifiers are enhanced, with an average improvement of up to 26% in performance.

Caracterização molecular dos genes PSO2 e PSO4 de Saccharomyces cerevisiae envolvidos na reparação do DNA : análise funcional e identificação de interações proteína-proteína

O alelo mutante termo-condicionalmente letal pso4-1 do gene PRP19, que codifica uma proteína associada ao spliceossoma, permitiu investigar a influência deste gene no processamento de pré-mRNA, reparação do DNA e esporulação. Fenótipos relacionados a genes portadores de introns foram correlacionados à temperatura. Sistemas repórteres para processamento de pré-mRNA e RT-PCR mostraram que eficiência de processamento de mRNA no mutante pso4-1 é inversamente correlacionada com a temperatura de crescimento. Uma mutação pontual, substituindo uma leucina por uma serina foi identificada dentro da região codificadora N-terminal do alelo Pso4-1 e afeta as propriedades bioquímicas de Pso4-1p. Entre 24 clones isolados utilizando o sistema doishíbridos, 7 foram identificados como partes dos genes RAD2, RLF2 e DBR1. RAD2 codifica uma endonuclease indispensável para a via reparação por excisão de nucleotídeos (NER), RLF2 codifica a subunidade maior do fator de montagem da cromatina I, cuja deleção resulta em sinsibilidade à radiação UVC, enquanto que DBR1 codifica uma enzima que atua sobre substratos de RNA na forma lariat, degradando estruturas lariat em introns durante o processamento de mRNA. A caracterização dos fenótipos após tratamentos com mutágenos em linhagens mutantes simples e duplos de rad2∆, rlf2∆ e pso4-1 mostraram sensibilidade aumentada para para mutantes rad2∆/pso4-1 e rlf2∆/pso4-1, sugerindo uma interferência funcional destas proteínas no processo dereparação do DNA em Saccharomyces cerevisiae. O mecanismo exato de reparação de pontes inter-cadeia (ICL) em S. cerevisiae não é ainda totalmente conhecido. Identificando novos fenótipos e isolando proteínas potencialmente capazes de interagir com Pso2p através da técnica do sistema dois-híbridos, foi possível extender a caracterização deste gene específica para reparação de pontes intercadeia. Tratamentos com acetaldeído (ACA), um metabólito natural da via glicolítica, foi mais tóxico a linhagem mutante pso2 em comparação com a linhagem selvagem e também capaz de induzir a expressão da fusão contendo o promotor de PSO2 à lacZ (PSO2-lacZ) por um fator comparável à tratamentos com outros agentes indutores de ICLs, indicando que o metabólito natural ACA pode causar danos do tipo ICL em S. cerevisiae. A utilização do sistema dois-híbridos permitiu isolar partes de proteínas codificadas por nove diferentes genes, entre eles a proteína quinase Pak1p, um supressor de mutações termosensíveis da DNA Polimerase alfa. Pak1p interage com a extremidade C-terminal conservada de Pso2p, uma região da proteína recentemente nomeada β-CASP entre v ortólogos conhecidos do gene PSO2. A integridade do domínio β-CASP é essencial para a reparaçãode DNA proficiente como demonstrado em ensaios de complementação com mutantes pso2 ∆. Comparação da sobrevivência após tratamento com agentes mutagênicos de simples mutantes pso2 ∆ e pak1 ∆ assim com o dulpo mutante pso2 ∆/pak1 ∆ revelaram que o gene PAK1 é necessário para reparação do DNA proficiente como na linhagem selvagem. A interação epistática dos dois alelos mutantes na linhagem duplo mutante sugere que Pak1p atua na mesma via de reparação a qual PSO2 pertence e que PAK1 constitui um novo locus envolvido na reparação do DNA em S. cerevisiae.

Os keynesianos neo-ricardianos e os pós-keynesianos: comentários a Amadeu e Dutt

O objetivo deste "paper" é tecer alguns comentários à leitura que Amadeo e Dutt apresentam em artigo publicado na Pesquisa e Planejamento Econômico, sobre duas vertentes do keynesianismo: a ner-ricardiana e a pós-keynesiana.

Motivação à prática regular de atividades físicas: um estudo sobre tenistas brasileiros infanto-juvenis

As pesquisas sobre as motivações que levam crianças e jovens à prática de atividades físicas e desportivas vêm recebendo crescente destaque na literatura desportiva. O presente trabalho teve por objetivo central explorar e avaliar um grupo de 6 (seis) das mais relevantes dimensões motivacionais associadas à prática regular de atividades físicas, que melhor descrevem os jovens tenistas brasileiros da faixa etária de 13 a 16 anos: Controle de Estresse, Saúde, Prazer, Competitividade, Sociabilidade e Estética. Mais especificamente, o estudo procurou verificar se há diferença estatisticamente significativa entre as dimensões motivacionais dos tenistas segundo as variáveis controladas: Sexo; Categorias (“até 14anos” e “até 16 anos”); Ranking (tenistas “integrantes do ranking” (IR) e tenistas “não integrantes do ranking” (NIR)); Experiência em competições (tenistas “estreantes em competições” (ER) e tenistas “não estreantes em competições” (NER)). Para tanto, aplicou-se o Inventário de Motivação à Prática Regular de Atividades Físicas (IMPRAF-126; Balbinotti, 2004). O IMPRAF-126 é respondido numa escala do tipo Likert de cinco pontos (1 – Isto me motiva pouquíssimo a 5 – Isto me motiva muitíssimo) para verificar, em valores nominais, as dimensões que mais motivam os tenistas à prática de atividades físicas regulares. O IMPRAF-126 foi aplicado em 226 jovens tenistas de ambos os sexos, com idades entre 13 e 16 anos. Todos os tenistas participam das competições promovidas pelas Federações Gaúcha e Catarinense de Tênis. Constatou-se que a dimensão que mais motiva os tenistas à prática regular de atividades físicas é o Prazer seguido por um grupo, indissociável estatisticamente, formado pela Competitividade e a Saúde. Seguem-se a este grupo a Sociabilidade, a Estética e o Controle de Estresse. Avaliando as motivações dos tenistas com as variáveis controladas, percebe-se que o Controle de Estresse motiva significativamente mais os tenistas do sexo masculino em comparação com as tenistas do sexo feminino. Tenistas da Categoria “até 16 anos” se motivam significativamente mais que os da categoria “até 14 anos” pela Sociabilidade. A Competitividade motiva significativamente mais aos tenistas IR do que aos tenistas NIR. Os tenistas NEC se motivam significativamente mais que os tenistas EC pela Competitividade. Os resultados deste estudo sugerem que a prática dos jovens tenistas brasileiros se origina predominantemente pelas suas motivações intrínsecas. Recomendamos que novos estudos com tenistas e com atletas de outros esportes, sejam realizados para aprofundar os conhecimentos sobre a motivação dos jovens à prática de atividades físicas.

A esquina do povo de alt’lam: aprimoramento racial e moral no movimento espírita natalense

A partir de um conjunto de dados obtidos em pesquisa, o texto deseja salientar uma face pouco examinada pelos estudiosos do espiritismo no Brasil: aquela dos grupos espíritas que cultuam extraterrestres. Examina a narrativa Rebelião de Lúcifer, escrita pelo médium potiguar Jan Val Ellam. Sustenta que esta atualiza noções importantes em relação a narrativas de salvação anteriores construídas pelo espiritismo. Destaca em especial o aprimoramento racial e moral através da reencarnação em distintos corpos humanos e em diferentes planetas. Examina-os a partir das noções de exílio planetário e de evolução racializada. Argumenta que Jan Val Ellam as atualiza através da categoria nativa Povo de Alt’Lam e através de bricolagem com outros campos semânticos, notadamente a ufologia e a fi cção científica

Modulação da expressão da proteína XPA em resposta ao tratamento com azul de metileno fotossensibilizado

Reactive oxygen species (ROS) are produced by aerobic metabolism and react with biomolecules, such as lipids, proteins and DNA. In high concentration, they lead to oxidative stress. Among ROS, singlet oxygen (1O2) is one of the main ROS involved in oxidative stress and is one of the most reactive forms of molecular oxygen. The exposure of some dyes, such as methylene blue (MB) to light (MB+VL), is able to generate 1O2 and it is the principle involved in photodynamic therapy (PDT). 1O2 e other ROS have caused toxic and carcinogenic effects and have been associated with ageing, neurodegenerative diseases and cancer. Oxidative DNA damage is mainly repaired by base excision repair (BER) pathway. However, recent studies have observed the involvement of nucleotide excision repair (NER) factors in the repair of this type of injury. One of these factors is the Xeroderma Pigmentosum Complementation Group A (XPA) protein, which acts with other proteins in DNA damage recognition and in the recruitment of other repair factors. Moreover, oxidative agents such as 1O2 can induce gene expression. In this context, this study aimed at evaluating the response of XPA-deficient cells after treatment with photosensitized MB. For this purpose, we analyzed the cell viability and occurrence of oxidative DNA damage in cells lines proficient and deficient in XPA after treatment with MB+VL, and evaluated the expression of this enzyme in proficient and complemented cells. Our results indicate an increased resistance to treatment of complemented cells and a higher level of oxidative damage in the deficient cell lines. Furthermore, the treatment was able to modulate the XPA expression up to 24 hours later. These results indicate a direct evidence for the involvement of NER enzymes in the repair of oxidative damage. Besides, a better understanding of the effects of PDT on the induction of gene expression could be provided

Papel da proteína de reparo XPC na regulação das proteínas de reparo APE1, OGG1 e PARP-1 em células humanas e de camundongos

studies using UV as a source of DNA damage. However, even though unrepaired UV-induced DNA damages are related to mutagenesis, cell death and tumorigenesis, they do not explain phenotypes such as neurodegeneration and internal tumors observed in patients with syndromes like Xeroderma Pigmentosum (XP) and Cockayne Syndrome (CS) that are associated with NER deficiency. Recent evidences point to a role of NER in the repair of 8-oxodG, a typical substrate of Base Excision Repair (BER). Since deficiencies in BER result in genomic instability, neurodegenerative diseases and cancer, it was investigated in this research the impact of XPC deficiency on BER functions in human cells. It was analyzed both the expression and the cellular localization of APE1, OGG1 e PARP-1, the mainly BER enzymes, in different NER-deficient human fibroblasts. The endogenous levels of these enzymes are reduced in XPC deficient cells. Surprisingly, XP-C fibroblasts were more resistant to oxidative agents than the other NER deficient fibroblasts, despite presenting the highest of 8-oxodG. Furthermore, subtle changes in the nuclear and mitochondrial localization of APE1 were detected in XP-C fibroblasts. To confirm the impact of XPC deficiency in the regulation of APE1 and OGG1 expression and activity, we constructed a XPC-complemented cell line. Although the XPC complementation was only partial, we found that XPC-complemented cells presented increased levels of OGG1 than XPC-deficient cells. The extracts from XPC-complemented cells also presented an elevated OGG1 enzimatic activity. However, it was not observed changes in APE1 expression and activity in the XPCcomplemented cells. In addition, we found that full-length APE1 (37 kDa) and OGG1- α are in the mitochondria of XPC-deficient fibroblasts and XPC-complemented fibroblasts before and after induction of oxidative stress. On the other hand, the expression of APE1 and PARP-1 are not altered in brain and liver of XPC knockout mice. However, XPC deficiency changed the APE1 localization in hypoccampus and hypothalamus. We also observed a physical interaction between XPC and APE1 proteins in human cells. In conclusion, the data suggest that XPC protein has a role in the regulation of OGG1 expression and activity in human cells and is involved mainly in the regulation of APE1 localization in mice. Aditionally, the response of NER deficient cells under oxidative stress may not be only associated to the NER deficiency per se, but it may include the new functions of NER enzymes in regulation of expression and cell localization of BER proteins

Análise da Expressão de APE1 após Estresse Oxidativo Induzido em Células Proficientes e Deficientes na Via de Reparo por Excisão de Nucleotídeos

Riboflavin is a vitamin very important in aerobic organisms, as a precursor of many coenzymes involved in the electron transporter chain. However, after photosensitization of riboflavin with UV or visible light, it generates reactive oxygen species (ROS), which can oxidize the DNA. The repair of oxidative lesions on DNA occurs through the base excision repair pathway (BER), where APE1 endonuclease plays a central role. On the other hand, the nucleotide excision repair pathway (NER) repairs helix-distorting lesions. Recently, it was described the participation of NERproteins in the repair of oxidative damage and in stimulation of repair function fromAPE1. The aim of this research was to evaluate the cytotoxic effects of photosensitized riboflavin (RF*) in cells proficient and deficient in NER, correlating with APE1 expression. For this propose, the cells were treated with RF* and it was performed the cell viability assay, extraction of whole proteins, cells fractionation, immunoblotting, indirect immunofluorescence and analysis of polymorphisms of BER gens. The results evidenced that cells deficient in XPA and CSB proteins were more sensitive to RF*. However, XPC-deficient cells presented similar resistance to MRC5- SV cells, which is proficient in NER. These results indicate that XPA and CSB proteins have an important role on repair of oxidative lesions induced by RF*. Additionally, it was evidenced that single nucleotide polymorphisms (SNPs) in BER enzymes may influence in sensitivity of NER-deficient cell lines. Concerning the APE1 expression, the results showed that expression of this protein after treatment with RF* only changed in XPC-deficient cells. Though, it was observed that APE1 is recruited and is bound to chromatin in MRC5-SV and XPA cells after treatment with RF*. The results also showed the induction of DNA damage after treatment with RF*, through the analysis of-H2AX, since the treatment promoted an increase of endogenous levels of this phosphorylated protein, which acts signaling double strand-break on DNA. On the other hand, in XPC-deficient cells, regardless of resistance of RF*, the endogenous levels of APE1 are extremely reduced when compared with other cell lines and APE1 is not bound to chromatin after treatment with RF*. These results conclude that RF* was able to induce cell death in NERdeficient cells, where XPA and CSB cells were more sensitive when compared with MRC5-SV and XPC-deficient cells. This last result is potentially very interesting, since XPC-deficient cell line presents low levels of APE1. Additionally, the results evidenced that APE1 protein can be involved in the repair of oxidative damage induced by RF*, because APE1 is recruited and bound strongly to chromatin after treatment.

Proteômica comparativa de linhagens celulares humanas expostas a estresse oxidativo induzido por riboflavina fotossensibilizada

Reactive oxygen species (ROS) are continuously generated and can be derived from cellular metabolism or induced by exogenous factors, in addition, have the capacity to damage molecules like DNA and proteins. BER is considered the main route of DNA damage oxidative repair, however, several studies have demonstrated the importance of the proteins participation of other ways to correct these injuries. NER enzymes deficiency, such as CSB and XPC, acting in the damage recognition step in the two subways this system influences the effectiveness of oxidative damage repair. However, the mechanisms by which cells deficient in these enzymes respond to oxidative stress and its consequences still need to be better understood. Thus, the aim of this study was to perform a proteomic analysis of cell lines proficient and deficient in NER, exposed to oxidative stress, in order to identify proteins involved, directly or not, in response to oxidative stress and DNA repair. For this, three strains of human fibroblasts, MRC5-SV, CS1AN (CSBdeficient) and XP4PA (XPC-deficient) were treated with photosensitized riboflavin and then carried out the differentially expressed proteins identification by mass spectrometry. From the results, it was observed in MRC5-SV increase expression in most of the proteins involved in cellular defense, an expected response to a normal cell line subjected to stress. CS1AN showed a response disjointed, it is not possible to establish many interactions between the proteins identified, may be one explanation for their sensitivity to treatment with riboflavin and other oxidants and increased cell death probably by induction of pro-apoptotic pathways. Already XP4PA showed higher expression of apoptosis-blocking proteins, as there was inhibition or reduced expression of others involved with the activation of this process, suggesting the activation of an anti-apoptotic mechanism in this lineage, which may help explain the high susceptibility to develop cancers in XPC individuals. These results also contribute to elucidate action mechanisms of NER in oxidative damage and the understanding of important routes in the oxidative stress correlation, repair and malignant tumors formation

Functional XPB/RAD25 redundancy in Arabidopsis genome: characterization of AtXPB2 and expression analysis

The xeroderma pigmentosum complementation group B (XPB) protein is involved in both DNA repair and transcription in human cells. It is a component of the transcription factor IIH (TFIIH) and is responsible for DNA helicase activity during nucleotide (nt) excision repair (NER). Its high evolutionary conservation has allowed identification of homologous proteins in different organisms, including plants. In contrast to other organisms, Arabidopsis thaliana harbors a duplication of the XPB orthologue (AtXPB1 and AtXPB2), and the proteins encoded by the duplicated genes are very similar (95% amino acid identity). Complementation assays in yeast rad25 mutant strains suggest the involvement of AtXPB2 in DNA repair, as already shown for AtXPB1, indicating that these proteins may be functionally redundant in the removal of DNA lesions in A. thaliana. Although both genes are expressed in a constitutive manner during the plant life cycle, Northern blot analyses suggest that light modulates the expression level of both XPB copies, and transcript levels increase during early stages of development. Considering the high similarity between AtXPB1 and AtXPB2 and that both of predicted proteins may act in DNA repair, it is possible that this duplication may confer more flexibility and resistance to DNA damaging agents in thale cress. (C) 2004 Elsevier B.V. All rights reserved.

Evolução do reparo por excisão de nucleotídeo

The phylogeny is one of the main activities of the modern taxonomists and a way to reconstruct the history of the life through comparative analysis of these sequences stored in their genomes aimed find any justification for the origin or evolution of them. Among the sequences with a high level of conservation are the genes of repair because it is important for the conservation and maintenance of genetic stability. Hence, variations in repair genes, as the genes of the nucleotide excision repair (NER), may indicate a possible gene transfer between species. This study aimed to examine the evolutionary history of the components of the NER. For this, sequences of UVRA, UVRB, UVRC and XPB were obtained from GenBank by Blast-p, considering 10-15 as cutoff to create a database. Phylogenetic studies were done using algorithms in PAUP programs, BAYES and PHYLIP package. Phylogenetic trees were build with protein sequences and with sequences of 16S ribosomal RNA for comparative analysis by the methods of parsimony, likelihood and Bayesian. The XPB tree shows that archaeal´s XPB helicases are similar to eukaryotic helicases. According to this data, we infer that the eukaryote nucleotide excision repair system had appeared in Archaea. At UVRA, UVRB and UVRC trees was found a monophyletic group formed by three species of epsilonproteobacterias class, three species of mollicutes class and archaeabacterias of Methanobacteria and Methanococci classes. This information is supported by a tree obtained with the proteins, UVRA, UVRB and UVRC concatenated. Thus, although there are arguments in the literature defending the horizontal transfer of the system uvrABC of bacteria to archaeabacterias, the analysis made in this study suggests that occurred a vertical transfer, from archaeabacteria, of both the NER genes: uvrABC and XPs. According the parsimony, this is the best way because of the occurrence of monophyletic groups, the time of divergence of classes and number of archaeabacterias species with uvrABC system

Time evolution of the Wigner function in discrete quantum phase space for a soluble quasi-spin model

The discrete phase space approach to quantum mechanics of degrees of freedom without classical counterparts is applied to the many-fermions/quasi-spin Lipkin model. The Wi:ner function is written for some chosen states associated to discrete angle and angular momentum variables, and the rime evolution is numerically calculated using the discrete von Neumnnn-Liouville equation. Direct evidences in the lime evolution of the Wigner function are extracted that identify a tunnelling effect. A connection with a SU(2)-based semiclassical continuous approach to the Lipkin model is also presented.