Quality Control in the DCL

 

Validated platforms

Prior to application, the available platforms are extensively validated in order to determine the performance characteristics, resulting in an error model per metabolite. Continuous quality control during application of these methods offers highly reliable data to be used for advanced data analysis.

Measurement Design

Measurement design includes the following controls

 

Randomization, Replication and Blocking

 

Technical samples for Quality Control

–          Quality Control (QC) Samples

• A sample representative for the study samples, i.e. a pooled study sample inserted at regular intervals during measurement runs.
• In the absence of analytical variation a QC sample  yields constant instrument response which makes it possible to monitor and correct analytical variation.

–          Replicate samples

• 10-15% of all running samples  measured in duplicate for independent  validation of QC based corrections

–          Calibration samples (blocks)

• As QC samples but spiked at different levels. Placed in 2-3 blocks at regular intervals
• Makes it possible to monitor linearity and if endogenous compounds are spiked allows for calculation of absolute concentrations

–          Blank samples

• Carry over and other blank effects

 

QC Report

The quality of all data is evaluated using an in-house developed QC reporting program. This allows us to assess the quality of each metabolite, allows the detection of injection or integration errors, finds outliers, and corrects data using blank subtraction,  internal standard corrections and QC correction algorithms.

 

 

Quality Control pipeline implemented and routinely applied for small and large studies

Example large studies:

IOP biomarkers for healthy aging (±3000  samples, 31 batches, measurement time ±28 weeks)

Dutch Twin Register (NTR) (±2500 samples, 28 batches, Measurement time ±28 weeks)

DIet Obesity and GENES (DIOGENES) (± 2000 plasma samples, Measurement time 26 weeks)