Data Management and Quality Control
Biomarker data management is a key step in every clinical bioinformatics study. Approximately 40% of the overall analysis time is dedicated to these tasks, which are crucial for the quality of the overall analysis.
Biomarker data management involves the integration and consistency checking of data collected from different domains. These can range from patient data provided by the clinics, to data generated in the lab by biological analysis platforms. The data management solutions developed by Quartz Bio allow to flexibly integrate data from different origins and formats. Consitency checks are then performed by analysts with a biotechnological knowledge of the data. All the data management processes are documented, auditable and reproducible. Robust data management ensures that the data analyses are developed on a solid ground.
Biomarker data management involves different steps:
- Parse of the different files and integration into a “biomarker database”
- Verification of the consistency of the identifiers (subject ids, visits, biomarkers, etc) across the different datasets
- Standardisation of the variable names
- Check of formats and variable ranges (compared to data specifications)
- Derivation of variables specific to the biomarker data analysis
Biomarker data are heterogeneous. Technology-specific quality controls must be applied in order to identify possible sources of bias that could impact the analysis. The biology behind the assay and the sample logistics in clinical settings are two of the most important factors that can impact the biomarker data quality.
Request a quote or information
Our experts are available to answer any questions or discuss a quote