Self-Organizing Maps as a New Tool for Classification of Plants at Lower Hierarchical Levels

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Automatic identification of terpenoid skeletons by feed-forward neural networks

Feed-forward neural networks (FFNNs) were used to predict the skeletal type of molecules belonging to six classes of terpenoids. A database that contains the (13)C NMR spectra of about 5000 compounds was used to train the FFNNs. An efficient representation of the spectra was designed and the constitution of the best FFNN input vector format resorted from an heuristic approach. The latter was derived from general considerations on terpenoid structures. (c) 2006 Elsevier B.V. All rights reserved.

Chemical evolution in the Asteraceae. The Oxidation-Reduction mechanism and production of secondary metabolites

This work describes the application of partial least squares (PLS) regression to variables that represent the oxidation data of several types of secondary metabolite isolated from the family Asteraceae. The oxidation states were calculated for each carbon atom of the involved compounds after these had been matched with their biogenetic precursor. The states of oxidation variations were named oxidation steps. This methodology represents a new approach to inspect the oxidative changes in taxa. Partial least square (PLS) regression was used to inspect the relationships among terpenoids, cournarins, polyacetylenes, and flavonoids from a data base containing approximately 27,000 botanical entries. The results show an interdependence between the average oxidation states of each class of secondary metabolite at tribe and sub tribe levels.

Modelo de Marcus, elementos de matriz de tunelamento, efeitos de solvente e controle nuclear em reações de transferência de elétrons

This paper aims to give an initial presentation to physics students of the area of electron transfer, addressing its major aspects. The subject is then presented in an introductory way, highlighting and discussing the key points. Primarily, the problem is approached in a simplified manner through a two-state system, and aspects of calculating the matrix-element are discussed. Then, the electron transfer problem, influenced by nuclear reaction coordinates, is addressed and treated by the Marcus theory. Time scales and the concepts of adiabatic and nonadiabatic phenomena are discussed in the context of electron transfer. The classical, semi-classical and quantum formalisms of for the nuclear reactions coordinates are presented, and aspects of solvents are also discussed. Finally, we conclude by examining some recent examples of problems discussed in the literature.

Characterization of monocarboxylate transporters (MCTs) expression in soft tissue sarcomas: distinct prognostic impact of MCT1 sub-cellular localization

Background: Soft tissue sarcomas (STSs) are a group of neoplasms, which, despite current therapeutic advances, still confer a poor outcome to half of the patients. As other solid tumors, STSs exhibit high glucose consumption rates, associated with worse prognosis and therapeutic response. As highly glycolytic tumors, we hypothesized that sarcomas should present an increased expression of lactate transporters (MCTs).Methods: Immunohistochemical expression of MCT1, MCT2, MCT4 and CD147 was assessed in a series of 86 STSs and the expression profiles were associated with patients' clinical-pathological parameters.Results: MCT1, MCT4 and CD147 were mainly observed in the plasma membrane of cancer cells (around 60% for MCTs and 40% for CD147), while MCT2 was conspicuously found in the cytoplasm (94.2%). Importantly, we observed MCT1 nuclear expression (32.6%). MCT1 and MCT4, alone or co-expressed with CD147 in the plasma membrane, were associated with poor prognostic variables including high tumor grade, disease progression and shorter overall survival. Conversely, we found MCT1 nuclear expression to be associated with low grade tumors and longer overall survival.Conclusions: The present work represents the first report of MCTs characterization in STSs. We showed the original finding of MCT1 expression in the nucleus. Importantly, opposite biological roles should be behind the dual sub-cellular localization of MCT1, as plasma membrane expression of MCT1 is associated with worse patients' prognosis, while nuclear expression is associated with better prognosis.

The genomes of two key bumblebee species with primitive eusocial organization

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)