Teaching Chemometrics with a Bioprocess: Analytical Methods Comparison Using Bivariate Linear Regression

Franco, Vanina G.; Mantovani, Víctor E.; Goicoechea, Héctor C.; Olivieri, Alejandro C.

doi:10.1007/s00897020596a

Cited by 8 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recommended test is the so-called elliptical joint confidence region (EJCR) test, which implies drawing the EJCR for the slope and intercept of the above linear plot, and checking whether the ideal point (slope = 1, intercept = 0) is included in the ellipse [22]. It should be noticed that for nonconstant prediction variance, a regression technique should be employed accounting for the fact that there are non-constant errors, such as weighted least-squares (WLS) or bilinear leastsquares (BLS), and not ordinary least-squares (OLS), as is usually done [23,24]. Otherwise, dramatically different conclusions might be obtained [24].…”

Section: Recovery Studiesmentioning

confidence: 99%

“…It should be noticed that for nonconstant prediction variance, a regression technique should be employed accounting for the fact that there are non-constant errors, such as weighted least-squares (WLS) or bilinear leastsquares (BLS), and not ordinary least-squares (OLS), as is usually done [23,24]. Otherwise, dramatically different conclusions might be obtained [24].…”

Section: Recovery Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Practical guidelines for reporting results in single- and multi-component analytical calibration: A tutorial

Olivieri

2015

Analytica Chimica Acta

Self Cite

270

View full text Add to dashboard Cite

Section: Recovery Studiesmentioning

confidence: 99%

Section: Recovery Studiesmentioning

confidence: 99%

Practical guidelines for reporting results in single- and multi-component analytical calibration: A tutorial

Olivieri

2015

Analytica Chimica Acta

Self Cite

270

View full text Add to dashboard Cite

“…These precisions can be considered using the different existing approaches for calculating regression coefficients and related statistical parameters that consider errors in both axes (Olivieri & Goicoechea, ; Riu & Rius, ). Franco, Mantovani, Goicoechea, and Olivieri () reported the dangers of not using a correct regression method for the comparison of analytical methods. In addition, the fact that in some cases the weighted least‐squares (WLS) and bivariate least‐squares (BLS) methods can produce similar results, but very different from those provided by OLS (ordinary linear regression), is discussed by these authors.…”

Section: Discussionmentioning

confidence: 99%

In vitro assays predicts mineral retention and apparent protein digestibility of different fish feed measured using a juvenileP. mesopotamicusmodel

et al. 2018

View full text Add to dashboard Cite

Feed chemical composition only provides information about nutrient content of a diet, but not its bioavailability. The aims were to determine mineral bioaccessibility and protein digestibility of three different fish feed, using in vitro methods and to correlate them with feed mineral retention and apparent protein digestibility measured by juvenile Piaractus mesopotamicus model, considering elliptical joint confidence region (EJCR) of weighted least-squares analysis. For in vivo assays, juvenile P. mesopotamicus were randomly stocked into conical tanks and fed during 38 days with a commercial diet, an experimental diet, or the same experimental diet added with phytase (0.2 g/kg). Total body mineral retention (mineral bioavailability indicator) and apparent protein digestibility of different diets were measured. For in vitro methods, mineral dialysability (mineral bioaccessibility indicator) and protein digestibility of the same diets were determined. No significant difference in morphometric and haematological markers were detected between diets. For iron, zinc and phosphorous bioavailability, and protein digestibility, weighted least-squares analysis showed the optimal point (1, 0) was included in EJCR, indicating in vivo and in vitro methods were comparable and correlated satisfactorily. In vitro methods were able to reproduce accurately the results obtained from traditional in vivo assays using juvenile P. mesopotamicus model. K E Y W O R D Sfish feed, mineral dialysability, mineral retention, protein digestibility

show abstract

“…Any prediction problem, whether systematic or random, will occur during the analytical procedure it will generate deviations from this linear behavior . For this reason, this correlation of results was examined through the elliptical joint confidence region (EJCR) method, which measures the acceptable range for the intercept and the slope simultaneously with a statistical confidence of 95%.…”

Section: Methodsmentioning

confidence: 99%

Multivariate calibration for the improvement of the quantification of cadmium in the presence of potassium as interferent by total reflection X‐ray fluorescence

2019

View full text Add to dashboard Cite

A multivariate calibration method was applied to improve the accuracy of the determination of cadmium in the presence of potassium by total reflection X‐ray fluorescence (TXRF). Due to TXRF's relatively low resolution, a particular interference caused by potassium at line Kα 3.31 keV as interference can compromise the cadmium determination at line Lα 3.13 keV, respectively. The method is based on the hypothesis that the application of a multivariate calibration method, for example, partial least squares, could reduce variations due to interference and, consequently, improve the selectivity and accuracy in Cd determination. In this work, this strategy was evaluated on the X‐ray fluorescence emission signals between 2.50 and 3.90 keV for a set of 26 different synthetic calibration mixtures and eight different mixtures for external calibration. Based on a certified reference material (Clay 2—CRM051) with high potassium levels, a significant improvement of accuracy for cadmium determination was observed, overcoming the problems associated with spectral interferences by potassium.

show abstract

Teaching Chemometrics with a Bioprocess: Analytical Methods Comparison Using Bivariate Linear Regression

Cited by 8 publications

References 2 publications

Practical guidelines for reporting results in single- and multi-component analytical calibration: A tutorial

Practical guidelines for reporting results in single- and multi-component analytical calibration: A tutorial

In vitro assays predicts mineral retention and apparent protein digestibility of different fish feed measured using a juvenileP. mesopotamicusmodel

Multivariate calibration for the improvement of the quantification of cadmium in the presence of potassium as interferent by total reflection X‐ray fluorescence

Contact Info

Product

Resources

About