In this work, three methods for pattern recognition, used as exploratory data analysis, are revisited. A brief review of principal component analysis, PCA, an unsupervised method, is provided. Next, the Mahalanobis distance and the confidence ellipses usually drawn around the scores samples are discussed. Fisher’s canonical variate analysis (a supervised methodology) is the second method revisited in this work. The third exploratory data analysis methodology addressed is ANOVA-PCA, which uses the analysis of variance to separate variations into main effects, interaction and noise followed by principal component analysis. Unlike the other two, ANOVA-PCA was proposed recently and is still not yet explored in all its capabilities. One advantage of this method is the possibility to calculate the variance of each of the effects involved in the experimental design. The mathematical bases of the three methods are discussed as well as examples are presented.

The analysis of volatiles is an alternative in the routine of quality control of fruits, since the volatile metabolites can change in the presence of microorganisms. A well-established method for extracting volatiles is the solid phase microextraction (SPME). This work aims to define some parameters (fiber coating, conditioning time and fiber exposure time) for an effective extraction. Whole papaya was inserted into a hermetically closed glass bottle and the system was left in conditioning for the volatiles accumulation. Then, extraction by SPME and subsequent desorption in a gas chromatograph was performed. Two fibers were evaluated: a 100 µm polydimethylsiloxane (PDMS) and a 65 µm polydimethylsiloxane/divinylbenzene (PDMS/DVB). The PDMS/DVB fiber extracted the highest number of volatiles. Central composite design (CCD) was performed to assess the effect of papaya conditioning time and the fiber exposure time on the response: number of detectable peaks. Both variables studied and their interactions were significant for the response. ANOVA showed that the quadratic terms and the lack of fit were not significant (p <0.05). A conditioning time of 10 min and exposure time of 30 min was sufficient for the detection of more than 100 compounds.

Abstract The real-time prediction in biorefinery industries has become essential. Models using partial least square regression (PLS) were developed to predict moisture, ash, volatile matter, fixed carbon and organic matter of coconut and coffee residues. In this study, 49 samples were collected and near infrared spectroscopy were applied to predict moisture, ash, volatile matter, fixed carbon and organic matter. For external validation 25% of the set samples were used. Moisture and volatile matter were predicted with coefficients of determination (R2cal) above 0.90, and standard errors (RSD) of the estimate of 14.4% and 2.26%, respectively. Models of ash and organic matter show R2cal > 0.77 and RSD values < 20.4%. For the external validation, the low deviations show the approximation between reference and predicted values and good prediction with R2pred > 0.70. All calibration models were acceptable for sample screening. This study demonstrates that PLS can be used to predict biomass composition of different species, with very low costs and time.

Neste trabalho, foram estudados os perfis cromatográficos dos compostos voláteis de 21 amostras de cafés comerciais brasileiros submetidos a diferentes processos industriais, i.e., descafeinados e com diferentes graus de torra (médio e extraforte). Os voláteis foram coletados por microextração em fase sólida (HS-SPME) e analisados por GC-FID e GC-MS. As matrizes de dados cromatográficos (perfis) obtidas, a partir dessas análises, foram exploradas por análise de componentes principais (PCA) e análise discriminante por quadrados mínimos parciais (PLS-DA). Inicialmente, os dados cromatográficos foram alinhados utilizando-se o algorítimo correlation optimized warping (COW). A PCA demonstrou a diferenciação dos cafés descafeinados em relação às outras classes de café, com ambas as fibras de SPME utilizadas. Essa separação, provavelmente, ocorreu em razão do desaparecimento de algumas espécies precursoras de voláteis durante o processo de descafeinação, tal como sacarose. Os modelos PLS-DA, para ambas as fibras testadas, PDMS/DVB e CX/PDMS, classificaram corretamente 100% das amostras, de acordo com o grau de torra (médio e extraforte); a principal diferença entre as classes foi a diferença de concentração de alguns compostos voláteis, tais como 2-metilfurano, 2-metilbutanal, 2,3-pentanodiona, pirazina, 2-pirrol carboxaldeído, furfural e 2-furan metanol.

The fingerprints of the volatile compounds of 21 commercial Brazilian coffee samples submitted to different industrial processing i.e. decaffeinated or different roasting degrees (traditional and dark) were studied. The volatiles were collected by headspace solid phase microextraction (HS-SPME) and analyzed by GC-FID and GC-MS. The chromatographic data matrices (fingerprints) obtained were explored by the principal component analysis (PCA) and partial least squares - discriminative analysis (PLS-DA). Initially the chromatographic profiles were aligned by the algorithm correlation optimized warping (COW). The PCA showed the discrimination of the decaffeinated coffees from the others with both the SPME fibres used. This separation probably occurred due to the loss of some volatile precursors during the decaffeination process, such as sucrose. For both the fibres tested, PDMS/DVB and CX / PDMS SPME, the PLS-DA models correctly classified 100% of the samples according to their roasting degree: (medium and dark), the main differences being the concentrations of some of the volatile compounds such as 2-methyl furan, 2-methylbutanal, 2,3-pentanedione, pyrazine, 2-carboxyaldehyde pyrrole, furfural and 2-furanmethanol.

A espectroscopia na região do infravermelho próximo (NIR) foi usada para determinar o teor de umidade em amostras de café cru. Foram construídos modelos de regressão usando o método dos mínimos quadrados parciais (PLS) com diferentes pré-tratamentos de dados e 157 espectros NIR coletados de amostras de café usando um acessório de reflectância difusa, na região entre 4500 e 10000 cm-1. Os espectros originais passaram por diferentes transformações e pré-tratamentos matemáticos, como a transformação Kubelka-Munk; a correção multiplicativa de sinal (MSC); o alisamento com SPLINE e a média móvel, e os dados foram escalados pela variância. O modelo de regressão permitiu determinar o teor de umidade nas amostras de café cru com erro quadrático médio de calibração (SEC) de 0,569 g.100 g -1; erro quadrático médio de validação de 0,298 g.100 g -1; coeficiente de correlação (r) 0,712 e 0,818 para calibração e validação, respectivamente; e erro relativo médio de 4,1% para amostras de validação.

Near infra-red reflectance (NIR) spectroscopy was used to measure the moisture content in raw coffee. Different models using partial least squares (PLS) with data pre-processing were used. Regression models were built with 157 spectra of the samples of raw coffee collected using a near infrared spectrometer with an accessory of diffuse reflectance, between 4500 and 10000 cm-1. The original NIR spectra went through different transformations and mathematical pre treatments, such as the Kubelka-Munk transformation; multiplicative signal correction (MSC); spline smoothing and movable average, and the data were scaled by variance. The regression model permitted the determination of the moisture content of the raw coffee samples with a standard error of calibration (SEC) = 0.569 g.100 g -1; standard error of validation = 0.298 g.100 g -1; correlation coefficient (r) 0.712 and 0.818 for calibration and validation, respectively, and average relative error of 4.1% for validation samples.

This work presents the VALIDATION SPREADSHEET, a public domain tool that can be used to evaluate the figures of merit for univariate analytical methods. A real example of BTEX determination in environmental samples is included to illustrate its use. The spreadsheet has been developed for Excel® and Open Office®, and is available on the internet at http//lqta.iqm.unicamp.br.

AIDS has the HIV as its etiological agent. Researches has been done to find new pharmacological agents to be used in therapy, because of problems of resistance and side effects. The HIV-integrase inhibitors are some of those new agents that are being studied. This updating focusses on the fundamental information about HIV and HIV-integrase and the main methods being used to develop these new drugs, with examples for each case.

The process of building mathematical models in quantitative structure-activity relationship (QSAR) studies is generally limited by the size of the dataset used to select variables from. For huge datasets, the task of selecting a given number of variables that produces the best linear model can be enormous, if not unfeasible. In this case, some methods can be used to separate good parameter combinations from the bad ones. In this paper three methodologies are analyzed: systematic search, genetic algorithm and chemometric methods. These methods have been exposed and discussed through practical examples.

Neste trabalho procurou-se correlacionar e agrupar amostras de água mineral provenientes de diferentes regiões do Estado de São Paulo a partir da análise exploratória dos valores de pH e teores de Ba, Ca, K, Mg, Na e V. Utilizando a análise de componentes principais (ACP) e a análise hierárquica de agrupamentos (AHA) foi possível avaliar a similaridade das amostras analisadas com a formação de grupos que estavam diretamente correlacionados com a procedência ou propriedades da água mineral. Em análise exploratória é utilizado termos como "scores" que fornecem a composição das componentes principais em relação aos objetos (amostras) e "loadings" fornecem essa mesma composição em relação às variáveis. Através dos gráficos de "scores" e "loadings" e dendogramas observou-se a formação de dois grupos distintos e a separação das amostras provenientes das cidades de Ibirá e Itú. O primeiro agrupamento observado é formado por amostras provenientes da região Serra da Mantiqueira (Águas da Prata, Campos do Jordão e Lindóia), enquanto que o segundo agrupamento compreende as amostras provenientes de cidades localizadas sobre o Aquífero Guarani (Águas de Santa Bárbara, Palmares Paulista, Presidente Prudente, São Carlos, São Simão e Santa Rosa do Viterbo). A separação das amostras provenientes de Ibirá ocorreu devido a predominância em sua composição mineral dos elementos V e Na, enquanto para as amostras procedentes de Itú foi observada maior contribuição dos elementos Ca, Ba e Mg. A aplicação da análise exploratória possibilitou correlacionar a composição química com a localização geográfica das amostras de água mineral, produzindo resultados úteis para o controle de qualidade e a construção de modelos de classificação ou previsão.

The correlation and grouping of mineral waters from different regions of São Paulo state was carried out using exploratory analysis of pH values and Ba, Ca, K, Mg, Na and V contents. By using principal components analysis (PCA) and hierarchical cluster analysis (HCA) was possible to evaluate the similarity of mineral water samples through group formation that was directly correlated with their sources. Employing scores, loadings and dendogram graphics was possible to observe the formation of two groups and the separation of Ibirá and Itú mineral water samples. The first group is composed by samples from Serra da Mantiqueira region (Águas da Prata, Campos do Jordão and Lindóia) and the other one is formed by samples from cities located above the Guarani Aquifer (Águas de Santa Bárbara, Palmares Paulista, Presidente Prudente, São Carlos, São Simão e Santa Rosa do Viterbo). The segregation of Ibirá samples was due to predominant presence of Na and V in their mineral composition, while for Itú samples it was observed higher contribution of Ba, Ca and Mg elements. The application of exploratory analysis enabled to correlate the chemical composition with the geographical source of the mineral water samples producing useful results for establishing of classification and prevision models.

Os teores de Ca, Cu, K, Mg, Na, P e S foram determinados simultaneamente em sucos e refrigerantes de uva utilizando espectrometria de emissão ótica com plasma de argônio induzido (ICP-OES) como técnica analítica. Os dados obtidos foram submetidos a dois métodos de análise exploratória: análise de componentes principais (PCA) e análise hierárquica de agrupamentos (HCA). Os resultados mostraram que o Na foi o principal elemento para discriminação entre as amostras, diferenciando principalmente sucos e refrigerantes. A adição de conservantes e o teor de sais minerais também foram aspectos relevantes para a discriminação dos grupos. A análise exploratória mostrou-se útil para o tratamento dos dados em questão, produzindo informações que dificilmente seriam visualizadas diretamente na matriz de dados.

The contents of Ca, Cu, K, Mg, Na, P and S were simultaneously determined in grape juices and soft drinks using ICP-OES. The experimental results were analyzed by methods of exploratory analysis: principal component analysis (PCA) and hierarchical cluster analysis (HCA). The results showed that Na was the principal element for discrimination among samples. Preservatives addition and mineral salts contents also were important to discriminate the clusters. Exploratory analysis was of great help in the data treatment giving information that would be too difficult to extract from the original data matrix.

Dois métodos de preparação de amostra para determinação de minerais em suco de uva foram comparados: extração com ácido clorídrico a frio com agitação e digestão em sistema fechado, sob pressão, empregando-se energia de microondas. As concentrações dos minerais foram obtidas empregando-se a técnica de Espectrometria de Emissão Óptica em Plasma Indutivamente Acoplado (ICP-OES). Os teores dos minerais Ca, P, Na, K, Mg, Zn, Fe, Mn e Cu não diferiram significativamente ao nível de 5% (Teste F) para as duas metodologias empregadas. A precisão e exatidão dos métodos foi avaliada usando o suco de uva. A metodologia de extração com ácido clorídrico foi empregada para a determinação dos teores de minerais em sucos processados de abacaxi, acerola, caju, goiaba, manga, maracujá e uva. Para facilitar a interpretação dos teores dos minerais encontrados, foram utilizadas as técnicas de Análise por Componentes Principais (PCA) e Análise Hierárquica por Agrupamento (HCA).

Two sample preparation methods for the determination of minerals from grape juice were investigated for the purpose of methodological evaluation: using hydrochloric acid by shaking, and digestion in closed system, under pressure, using microwave power. The concentrations of minerals were evaluated by the technique of Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES). The minerals contents of Ca, P, Na, K, Mg, Zn, Fe, Mn, and Cu did not differ significantly at the level of 5% (it Tests F) for both methodologies used. The precision and accuracy of both methodologies were evaluated using grape juice. The extraction methodology with hydrochloric acid was used for the determination of mineral amounts in processed juices from pineapple, acerola, cashew, guava, mango, passion fruit and grape. To enhance the interpretation of the amounts of minerals, from a nutritional point of view, techniques of Principal Components Analysis (PCA) and Hierarchic Cluster Analysis were used (HCA).