RESUMO
Paracoccidioides brasiliensis, the etiologic agent of paracoccidioidomycosis, is a dimorphic fungus which is found as mycelia (M) at 26 degrees C and as yeasts (Y) at 37 degrees C, or after the invasion of host tissues. Although the dimorphic transition in P. brasiliensis and other dimorphic fungi is an essential step in the establishment of infection, the molecular events regulating this process are yet poorly understood. Since the differential gene expression is a well-known mechanism which plays a central role in the dimorphic transition as well as in other biological process, in this work we describe the identification and characterization of two differentially expressed P. brasiliensis hydrophobin cDNAs (Pbhyd1 and Pbhyd2). Hydrophobins are small hydrophobic proteins related to a variety of important functions in fungal biology, including cell growth, development, infection, and virulence. These two hydrophobin genes are present as single copy in P. brasiliensis genome and Northern blot analysis revealed that both mRNAs are mycelium-specific and highly accumulated during the first 24 h of M to Y transition.
Assuntos
Proteínas Fúngicas/genética , Proteínas Fúngicas/fisiologia , Regulação Fúngica da Expressão Gênica , Micélio/crescimento & desenvolvimento , Paracoccidioides/crescimento & desenvolvimento , Paracoccidioides/genética , Regiões 3' não Traduzidas , Regiões 5' não Traduzidas , Sequência de Aminoácidos , Sequência de Bases , Cisteína/genética , DNA Complementar/química , DNA Complementar/isolamento & purificação , DNA Fúngico/química , Proteínas Fúngicas/química , Íntrons , Modelos Moleculares , Dados de Sequência Molecular , Micélio/genética , Paracoccidioides/citologia , Filogenia , RNA Fúngico/análise , RNA Mensageiro/análise , Homologia de Sequência de AminoácidosRESUMO
Paracoccidioides brasiliensis is a pathogenic fungus that undergoes a temperature-dependent cell morphology change from mycelium (22 degrees C) to yeast (36 degrees C). It is assumed that this morphological transition correlates with the infection of the human host. Our goal was to identify genes expressed in the mycelium (M) and yeast (Y) forms by EST sequencing in order to generate a partial map of the fungus transcriptome. Individual EST sequences were clustered by the CAP3 program and annotated using Blastx similarity analysis and InterPro Scan. Three different databases, GenBank nr, COG (clusters of orthologous groups) and GO (gene ontology) were used for annotation. A total of 3,938 (Y = 1,654 and M = 2,274) ESTs were sequenced and clustered into 597 contigs and 1,563 singlets, making up a total of 2,160 genes, which possibly represent one-quarter of the complete gene repertoire in P. brasiliensis. From this total, 1,040 were successfully annotated and 894 could be classified in 18 functional COG categories as follows: cellular metabolism (44%); information storage and processing (25%); cellular processes-cell division, posttranslational modifications, among others (19%); and genes of unknown functions (12%). Computer analysis enabled us to identify some genes potentially involved in the dimorphic transition and drug resistance. Furthermore, computer subtraction analysis revealed several genes possibly expressed in stage-specific forms of P. brasiliensis. Further analysis of these genes may provide new insights into the pathology and differentiation of P. brasiliensis.
Assuntos
Etiquetas de Sequências Expressas , Genoma Fúngico , Paracoccidioides/genética , Sequência de Bases , Brasil , Análise por Conglomerados , DNA Fúngico/química , DNA Fúngico/genética , Humanos , Dados de Sequência Molecular , Reação em Cadeia da Polimerase , Análise de Sequência de DNA , Transcrição GênicaRESUMO
This paper presents a new scheme for training MLPs which employs a relaxation method for multi-objective optimization. The algorithm works by obtaining a reduced set of solutions, from which the one with the best generalization is selected. This approach allows balancing between the training error and norm of network weight vectors, which are the two objective functions of the multi-objective optimization problem. The method is applied to classification and regression problems and compared with Weight Decay (WD), Support Vector Machines (SVMs) and standard Backpropagation (BP). It is shown that the systematic procedure for training proposed results on good generalization neural models, and outperforms traditional methods.
