Valid machine learning algorithms for multiparameter methods

Uhlig, Steffen; Colson, Bertrand; Hettwer, Karina; Simon, Kirsten; Uhlig, Carsten; Wittke, Stefan; Stoyke, M.; Gowik, Petra

doi:10.1007/s00769-019-01384-w

“…Calculations for median, mean value, and standard deviation were calculated in Excel® (Office 2019). The t test (p value < 0.05) and the PCA were performed in markerview (ABSciex, Darmstadt, Germany) as described by Lasch et al (2019) and Uhlig et al (2019).…”

Section: Discussionmentioning

confidence: 99%

“…Finally, principal component analysis (PCA) was used to evaluate the discriminative power by distinguishing the seven subgroups (skin, tendon, breast meat, scraped off meat, thighs, and offal vs intervertebral disc/cartilage) on the basis of the marker ions pattern. In general, only the two axes with the greatest variability are retained in PCA, and the result is displayed as a 2-D scatter plot (Lasch et al 2019;Uhlig et al 2019).…”

Section: Definition Of Specific Marker Ions (Markerview)mentioning

confidence: 99%

Detection of Mechanically Separated Meat from Chicken in Sausages and Cold Meat by Targeted LC–MS/MS Analysis

Wilhelm

¹

,

Hofsommer²,

Wittke

³

2022

Self Cite

View full text Add to dashboard Cite

The use of mechanically separated meat (MSM) from poultry in meat and sausage products is subject to declaration. Current methods such as microscopy or calcium analysis have proven to be insufficient to ensure the specific detection of MSM in meat and sausage products. When using MSM during production, intervertebral disc and cartilage specific proteins from chicken unavoidably end up in the sausages. Thus, a pseudo-MRM-LC–MS/MS-based assay was developed and validated, which uses intervertebral disc and cartilage specific peptides to detect MSM in meat and sausage products. All five marker peptides were assigned to collagen II alpha 1 which makes up a large part of the proteome of intervertebral discs and cartilage. In order to evaluate the validity of the methodology, a total of 23 positive controls (MSM content 5–90%) and a total of 19 negative controls were examined in a blinded study. After unblinding, 22 of 23 positive controls were correctly classified. Only one self-produced sample with 5% MSM was declared as a negative case (overall sensitivity 96%). In contrast, all negative controls were correctly classified as negative (specificity 100%). In summary, the LC–MS/MS assay allowed the specific detection of MSM in real samples with unknown composition down to 10% MSM in the meat content. Graphical abstract

show abstract

“…MacNeil et al aptly pointed out that validation, just like beauty, is in the eyes of the beholder [32]. In the context of NTMs, validation often refers to "model validation" (dry lab) [13,15], which can be done using techniques such as "nested cross-validation", or "kfold validation" (details of these techniques are described elsewhere [33,34]). Thus, it is necessary here to revisit MV definitions and review what can be applicable to NTM validation.…”

Section: Validation Definitionsmentioning

confidence: 99%

Essential Terminology and Considerations for Validation of Non-targeted Methods

Nichani¹,

Uhlig²,

Stoyke³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

Through its suggestive name, non-targeted methods (NTMs) do not aim at a predefined "needle in the haystack". Instead, they exploit all the constituents of the haystack. This new form of analytical methods is increasingly finding applications in food and feed testing. However, the concepts, terms, and considerations related to this burgeoning field of analytical testing needs to be propagated for the benefit of ones associated in academic research, commercial development, and official control. This paper addresses the frequently asked questions around notations and terminologies surrounding NTMs. The widespread development and adoption of these methods also necessitates the need to develop approaches to NTM validation, i.e., evaluating the performance characteristics of a method to determine if it is fit-for-purpose. This work aims to provide a roadmap to approaching NTM validation. In doing so, the paper deliberates on the different considerations that influence the approach to validation and provides suggestions thereof.

show abstract

“…MacNeil et al aptly pointed out that validation, just like beauty, is in the eyes of the beholder [30]. In the context of NTMs, validation often refers to "model validation" (dry lab) [13,15], which can be done using techniques such as "nested cross-validation", or "kfold validation" (details of these techniques are described elsewhere [31,32]). Thus, it is necessary here to revisit MV definitions and review what can be applicable to NTM validation.…”

Section: Validation Definitionsmentioning

confidence: 99%

Towards a Mathematical Framework for Method Validation and Performance Characteristics of Non-targeted Methods for Food Authenticity

Nichani¹,

Uhlig²,

Stoyke³

et al. 2021

Preprint

Self Cite

0

View full text Add to dashboard Cite

Through its suggestive name, non-targeted methods (NTMs) do not aim at a predefined "needle in the haystack". Instead, they exploit all the constituents of the haystack. This new form of analytical methods is increasingly finding applications in food and feed testing. However, the concepts, terms, and considerations related to this burgeoning field of analytical testing needs to be propagated for the benefit of ones associated in academic research, commercial development, and official control. This paper addresses the frequently asked questions around notations and terminologies surrounding NTMs. The widespread development and adoption of these methods also necessitates the need to develop approaches to NTM validation, i.e., evaluating the performance characteristics of a method to determine if it is fit-for-purpose. This work aims to provide a roadmap to approaching NTM validation. In doing so, the paper deliberates on the different considerations that influence the approach to validation and provides suggestions thereof.

show abstract

Valid machine learning algorithms for multiparameter methods

Cited by 10 publications

References 6 publications

Detection of Mechanically Separated Meat from Chicken in Sausages and Cold Meat by Targeted LC–MS/MS Analysis

Detection of Mechanically Separated Meat from Chicken in Sausages and Cold Meat by Targeted LC–MS/MS Analysis

Essential Terminology and Considerations for Validation of Non-targeted Methods

Towards a Mathematical Framework for Method Validation and Performance Characteristics of Non-targeted Methods for Food Authenticity

Contact Info

Product

Resources

About