Next Generation Sequencing (NGS) relies on several quite distinct steps, all with potential effect on the patient treatment. The wet lab process, comprising of specimen handling, DNA isolation and subsampling (capture or amplification), library preparation has its own challenges, some of which can be observed by looking at quality metrics of the resulting bioinformatic files. The sequencing instrument with its own potential pitfalls creates the shift into the digital space of NGS. The raw sequence data output then enters the bioinformatics process to eventually call the variants.
Bioinformatics are quite complex when it comes to the large volumes of NGS data in contrast to standard molecular tests. Algorithms evolve and so do the pipelines for different types of genetic tests. This high complexity, which comes on top of the permanent changes in the wet lab, creates a need for good monitoring procedures with special hindsight to validation and verification.
Validation and verification of laboratory tests requires demonstrated evidence of high consistency of performance towards true observation. How to gather this evidence for different variant types, different sequencing conditions, or different chromosomal regions, and in sufficient amounts to reach statistical significance?
In a follow-up webinar to the first (link below), Euformatics’ Chief Scientific Officer, Christophe Roos will dive a bit deeper into validation.