The use of correlation, association and regression to analyse processes and products

In recent years, as a result of the process analytical technology initiative of the US Food and Drug Administration, many different works have been carried out on direct and in situ monitoring of critical parameters for mammalian cell cultures by Raman spectroscopy and multivariate regression techniques. However, despite interesting results, it cannot be said that the proposed monitoring strategies, which will reduce errors of the regression models and thus confidence limits of the predictions, are really optimized. Hence, the aim of this article is to optimize some critical steps of spectroscopic acquisition and data treatment in order to reach a higher level of accuracy and robustness of bioprocess monitoring. In this way, we propose first an original strategy to assess the most suited Raman acquisition time for the processes involved. In a second part, we demonstrate the importance of the interbatch variability on the accuracy of the predictive models with a particular focus on the optical probes adjustment. Finally, we propose a methodology for the optimization of the spectral variables selection in order to decrease prediction errors of multivariate regressions. © 2017 American Institute of Chemical Engineers Biotechnol. Prog., 33:308-316, 2017.

show abstract

“…A 95% confidence interval is defined in Eq. considering

{\overset{y}{true}}_{i}

as the predicted concentration of sample i :

{\overset{y}{true}}_{i} {- 2 \times RMSEP < y}_{i} < {\overset{y}{true}}_{i} + 2 \times RMSEP

…”

Section: Resultsmentioning

confidence: 99%

Mammalian cell culture monitoring using in situ spectroscopy: Is your method really optimised?

André

Lagresle

Hannas³

et al. 2017

Biotechnology Progress

View full text Add to dashboard Cite

show abstract

“…The OLSR models were probably suffering from overfitting. According to Cozzolino (2014), if too many independent variables are used to model a response, the solution can become overfitted – as the model will become very dependent on the data set and will give poor prediction results. This demerit of OLSR is where the advantage of PCR and PLSR modeling lies.…”

Section: Resultsmentioning

confidence: 99%

Optimization and prediction of antioxidant properties of a tea‐ginger extract

Makanjuola

Enujiugha

Omoba

et al. 2015

Food Science & Nutrition

View full text Add to dashboard Cite

A response surface approach was used to investigate the effects of temperature, concentration, and time on the antioxidant properties (total flavonoid (TF), total phenol (TP), peroxide scavenging activity (PS), iron chelating activity (IC), DPPH radical-scavenging ability (DPPH), ABTS assay (ABTS)) of aqueous extract of tea-ginger (2:1) powder. Color indices, pH, and redox potential of the tea-ginger powder were also measured and used as independent variables for the prediction of antioxidant properties of the extract using ordinary least square (OLSR), principal component (PCR), and partial least square (PLSR) regression. The R2 values for TP, TF, ABTS, and PS response surface models were 0.8873, 0.9639, 0.6485, and 0.5721, respectively. The OLSR, PCR, and PLSR were able to provide predictive models for DPPH, TP, and TF of the tea-ginger extract (P < 0.05). The PLSR gave the most parsimonious model with an R2 of 0.851, 0.736, and 0.905 for DPPH, TP, and TF, respectively.

show abstract

“…It relies heavily on the use of mathematics and statistics in interpreting the data to provide definitive results. While chemometrics was first mooted back in 1995, it took almost twenty years for spectroscopic instrumentation to be fitted with effective and reproducible software tools to allow researchers to incorporate chemometrics into the processing of their spectral data to give absolute verifiable identification and quantitation of chemical components that would otherwise have been missed [66][67][68].…”

Section: Spectroscopic Methods and The Use Of Chemometricsmentioning

confidence: 99%

A Brief History of Whiskey Adulteration and the Role of Spectroscopy Combined with Chemometrics in the Detection of Modern Whiskey Fraud

et al. 2020

Self Cite

View full text Add to dashboard Cite

Food fraud and adulteration is a major concern in terms of economic and public health. Multivariate methods combined with spectroscopic techniques have shown promise as a novel analytical strategy for addressing issues related to food fraud that cannot be solved by the analysis of one variable, particularly in complex matrices such distilled beverages. This review describes and discusses different aspects of whisky production, and recent developments of laboratory, in field and high throughput analysis. In particular, recent applications detailing the use of vibrational spectroscopy techniques combined with data analytical methods used to not only distinguish between brand and origin of whisky but to also detect adulteration are presented.

show abstract

The use of correlation, association and regression to analyse processes and products

Cited by 6 publications

References 44 publications

Mammalian cell culture monitoring using in situ spectroscopy: Is your method really optimised?

Mammalian cell culture monitoring using in situ spectroscopy: Is your method really optimised?

Optimization and prediction of antioxidant properties of a tea‐ginger extract

A Brief History of Whiskey Adulteration and the Role of Spectroscopy Combined with Chemometrics in the Detection of Modern Whiskey Fraud

Contact Info

Product

Resources

About