FoodScreener Banner3

Statistical Analysis

Targeted and Non-Targeted Multi Marker Analysis

Targeted and Non-Targeted Multi Marker Analysis
Verification models are used for the non-targeted analysis of even unknown compounds

SGF Profiling delivers a standard targeted multi-marker analysis incorporating absolute quantification of:

  • Sugars (glucose, fructose, sucrose)
  • Main fruit acids (citric acid, malic acid, isocitric acid, quinic acid)
  • Perishable indicators (ethanol, fumaric acid, lactate, HMF)
  • Process control parameters (galacturonic acid, phlorin)

In addition the technique allows a non-targeted multi marker approach that is based on the simultaneous assessment of concentration deviations of hundreds of compounds. In contrast to targeted standard analytical routines, it can detect the appearance of unexpected ingredients to enable the detection of unknown fraud.
 

Spectroscopic Database

The screening is based on an extensive spectroscopic database that includes thousands of NMR spectra from mainly authentic juices. Currently the data base includes about 40 different fruit types from more than 50 production sites worldwide. In addition, the database also provides access to over hundreds of small molecule compounds for further analysis of unknown ingredients.
 

Push-Button Routine

SGF Profiling is a fully automated push-button routine that needs no interaction by the operator. From sample bar code registration, preparation and handling, to data acquisition and statistical evaluation, all steps are under the control of SampleTrack™, Bruker’s laboratory information system.
 

Reporting

Quantification

SGF Profiling delivers the absolute concentrations of ~30 juice ingredients that are critical for juice assessment. The values are compared to reference standards and deviations indicate characteristic quality issues, such as the addition of sugar.

Sample Classification

Sample classification helps to further differentiate between similar fruit types such as orange, blood-orange and mandarin. More specialized models can even distinguish between direct juice and rediluted juice, and enable to origin determination.

Verification

After sample classification the uni- and multi-variate verification delivers even more information, such as unexpected deviation from the reference group.

Regression Analyses

The regression analysis, based on training data sets, evaluates additional parameters such as titratable acids.

Estimation of Fruit Content

Included in the final report is also an estimation of the fruit content

Juice Screener SGF Profiling Report
SGF Profiling archives all results in the form of a standardized sample quality report