Artículos Q&Chttp://hdl.handle.net/10259/42502024-03-29T08:14:05Z2024-03-29T08:14:05ZDesign of hybrids for the minimum sum-of-squares clustering problemPacheco Bonrostro, JoaquínValencia García, Olgahttp://hdl.handle.net/10259/84772024-01-26T01:05:26Z2003-06-01T00:00:00ZDesign of hybrids for the minimum sum-of-squares clustering problem
Pacheco Bonrostro, Joaquín; Valencia García, Olga
A series of metaheuristic algorithms is proposed and analyzed for the non-hierarchical clustering problem under the criterion of minimum sum-of-squares clustering. These algorithms incorporate genetic operators and local search and tabu search procedures. The aim is to obtain quality solutions with short computation times. A series of computational experiments has been performed. The proposed algorithms obtain better results than previously reported methods, especially with a small number of clusters.
2003-06-01T00:00:00ZPrincipal component regression that minimizes the sum of the squares of the relative errors: Application in multivariate calibration modelsValencia García, OlgaOrtiz Fernández, Mª CruzSarabia Peinador, Luis Antoniohttp://hdl.handle.net/10259/82812024-01-11T01:05:24Z2021-01-01T00:00:00ZPrincipal component regression that minimizes the sum of the squares of the relative errors: Application in multivariate calibration models
Valencia García, Olga; Ortiz Fernández, Mª Cruz; Sarabia Peinador, Luis Antonio
Relative errors are typically used in chemometrics to evaluate the performance of a multivariate predictive model. However, these models are not obtained through the criterion of minimizing relative errors, as would be expected in a model whose response is the concentration of an analyte. There are no studies in chemometrics on the use of a principal component regression that minimizes the sum of the squares of the relative errors. This work proposes a model, which serves this purpose. The suggested model, wPCR, has been applied to 7 datasets with 12 predicted responses, 10 of which are multivariate calibrations of analytes in complex mixtures based on instrumental signals coming from various analytical techniques. As PCR and wPCR are methods seeking to optimize different criteria, each one achieves a better performance with respect to its own criterion. Therefore, the new model wPCR leads to better results insofar as the relative errors are considered, especially for the smallest responses. In this sense, the wPCR model also outperforms PCR with logarithmic transformation of the response (logPCR). In addition, as for the performance of the method using Joint Confidence Regions for the intercept and the slope of the accuracy line, it is shown that the application of wPCR does not introduce bias, neither constant nor proportional for the models built, nor a systematic alteration of the achievable accuracy.
2021-01-01T00:00:00ZA new efficient sample pooling procedure for qualitative screening analysis. Application to the detection of salmonella spp and nut allergen by PCRValencia García, OlgaSarabia Peinador, Luis AntonioOrtiz Fernández, Mª Cruzhttp://hdl.handle.net/10259/82742024-01-11T01:05:13Z2023-01-01T00:00:00ZA new efficient sample pooling procedure for qualitative screening analysis. Application to the detection of salmonella spp and nut allergen by PCR
Valencia García, Olga; Sarabia Peinador, Luis Antonio; Ortiz Fernández, Mª Cruz
Foodborne pathogens and allergens are a major concern for public health that determines food safety policy, and their screening could be improved using pooled samples.
The purpose of this paper is to propose a cost-effective and accurate pooling strategy for the unambiguous identification of foodborne pathogens and allergens, enabling laboratories to redouble testing capacity while saving time and making optimal use of resources.
Although a variety of pooling algorithms have been used in different fields, the strategy suggested here is a logical analysis of sample pooling aimed at qualitative analytical screening problems. It involves, on the one hand, a design matrix to make the pooled samples, which is a supersaturated-based design, particularly, a half-fraction of a Plackett-Burman. On the other hand, a logical (non-algebraic) modeling of the problem, as well as a logical procedure for the identification of the original positive samples, is included.
Regarding the efficiency of the proposal, it is higher than that of other pooling algorithms, with an expected number of tests per individual sample ranging between 0.10, for a prevalence below 1%, and 0.59, for a prevalence above 10%. In terms of accuracy, the pooling sensitivity reaches 0.9697 for a sensitivity of the analytical test about 0.99, while pooling specificity ranges from 0.9872 to 0.9999, provided that the sensitivity and specificity of the analytical test are equal to or greater than 0.90.
This pooling strategy has been applied to the detection of salmonella spp and nut allergen with complete identification of the contaminated samples. Furthermore, based on the EU reported food samples contaminated with salmonella, a comprehensive comparison between individual sampling and the applied pooling strategy has been conducted. For the specific case of salmonella, detailed calculations have been made on the expected efficiency gains induced by this pooling methodology for different types of food samples monitored in the EU.
2023-01-01T00:00:00ZLogical analysis of sample pooling for qualitative analytical testingSarabia Peinador, Luis AntonioValencia García, OlgaOrtiz Fernández, Mª Cruzhttp://hdl.handle.net/10259/77572023-07-14T00:05:16Z2023-09-01T00:00:00ZLogical analysis of sample pooling for qualitative analytical testing
Sarabia Peinador, Luis Antonio; Valencia García, Olga; Ortiz Fernández, Mª Cruz
When the prevalence of positive samples in a whole population is low, the pooling of samples to detect them has been widely used for epidemic control. However, its usefulness for applying analytical screening procedures in food safety (microbiological or allergen control), fraud detection or environmental monitoring is also evident.
The expected number of tests per individual sample that is necessary to identify all ‘positives’ is a measure of the efficiency of a sample pooling strategy. Reducing this figure is key to an effective use of available resources in environmental control and food safety. This reduction becomes critical when the availability of analytical tests is limited, as the SARS-CoV-2 pandemic showed.
The outcome of the qualitative analytical test is binary. Therefore, the operation governing the outcome of the pooled samples is not an algebraic sum of the individual results but the logical operator
(‘or’ in natural language). Consequently, the problem of using pooled samples to identify positive samples naturally leads to proposing a system of logical equations. Therefore, this work suggests a new strategy of sample pooling based on: i) A half-fraction of a Placket-Burman design to make the pooled samples and ii) The logical resolution, not numerical, to identify the positive samples from the outcomes of the analysis of the pooled samples.
For a prevalence of ‘positive’ equal to 0.05 and 10 original samples to be pooled, the algorithm presented here results in an expected value per individual equal to 0.37, meaning a 63% reduction in the expected number of tests per individual sample.
With sensitivities and specificities of the analytical test ranging from 0.90 to 0.99, the expected number of tests per individual ranges from 0.332 to 0.416, always higher than other pooled testing algorithms. In addition, the accuracy of the algorithm proposed is better or similar to that of other published algorithms, with an expected number of hits ranging from 99.16 to 99.90%.
The procedure is applied to the detection of food samples contaminated with a pathogen (Listeria monocytogenes) and others contaminated with an allergen (Pistachio) by means of Polymerase Chain Reaction, PCR, test.
2023-09-01T00:00:00Z