Computing the Kekulé structure count for alternant hydrocarbons

A fast computer algorithm brings computation of the permanents of sparse matrices, specifically, molecular adjacency matrices. Examples and results are presented, along with a discussion of the relationship of the permanent to the Kekulé structure count. A simple method is presented for determining the Kekulé structure count of alternant hydrocarbons. For these hydrocarbons, the square of the Kekulé structure count is equal to the permanent of the adjacency matrix. In addition, for alternant structures the adjacency matrix for N atoms can be written in such a way that only an N/2 × N/2 matrix need be evaluated. The Kekulé structure count correlates with topological indices. The inclusion of the number of cycles improves the fit. When comparing with previous results, the variance decreases 74%. The calculated standard heat of formation correlates with the logarithm of the Kekulé structure count. This heat increments 349 kJ/mol each time the Kekulé structure count increases by one order of magnitude. © 2002 Wiley Periodicals, Inc. Int J Quantum Chem, 2002     

Variable selection for multivariate calibration using a genetic algorithm: prediction of additive concentrations in polymer films from Fourier transform-infrared spectral data

Variable selection using a genetic algorithm is combined with partial least squares (PLS) for the prediction of additive concentrations in polymer films using Fourier transform-infrared (FT-IR) spectral data. An approach using an iterative application of the genetic algorithm is proposed. This approach allows for all variables to be considered and at the same time minimizes the risk of overfitting. We demonstrate that the variables selected by the genetic algorithm are consistent with expert knowledge. This very exciting result is a convincing application that the algorithm can select correct variables in an automated fashion.     

Evaluation of healthy and sensory indexes of sweetened beverages using an electronic tongue

Overconsumption of sugar-sweetened beverages may increase the risk of health problems and so, the evaluation of their glycemic load and fructose-intolerance level is essential since it may allow establishing possible relations between physiologic effects of sugar-rich beverages and health. In this work, an electronic tongue was used to accurately classify beverages according to glycemic load (low, medium or high load) as well to their adequacy for people suffering from fructose malabsorption syndrome (tolerable or not): 100% of correct classifications (leave-one-out cross-validation) using linear discriminant models based on potentiomentric signals selected by a meta-heuristic simulated annealing algorithm. These results may be partially explained by the electronic tongue’s capability to mimic the human sweetness perception and total acid flavor of beverages, which can be related with glycemic load and fructose-intolerance index. Finally, the E-tongue was also applied to quantify, accurately, healthy and sensory indexes using multiple linear regression models (leave-one-out cross-validation: R_adj > 0.99) in the following dynamic ranges: 4.7 < glycemic load ≤ 30; 0.4 < fructose intolerance index ≤ 1.5; 32 < sweetness perception < 155; 1.3 < total acid flavor, g L⁻¹ < 8.3; and, 5.8 < well-balanced flavor ≤ 74. So, the proposed electronic tongue could be used as a practical, fast, low-cost and green tool for beverage’s healthy and sensory evaluation.