Skip to content

Curating data for PK-DB

PK-DB Data logo

PK-DB is an open pharmacokinetics database for collecting, checking and reusing data from clinical and preclinical studies. Pharmacokinetics describes how the body absorbs, distributes, metabolizes and eliminates substances. PK-DB brings measurements from individual publications together with the experimental context needed to interpret and compare them.

What PK-DB contains

Each study connects information about the subjects, the treatment they received and the resulting measurements:

  • Groups and individuals: population size, age, sex, body weight, health status and other characteristics relevant to the study.
  • Interventions: administered substances, doses, routes and dosing schedules.
  • Measurements: concentration–time data and pharmacokinetic parameters such as clearance, half-life, area under the curve and maximum concentration.
  • Source information: references, figures, tables, units, reported variability and descriptions that connect the curated records to the publication.

This context supports comparison across studies, meta-analysis and pharmacokinetic modeling. Data can be explored in the web interface or accessed programmatically through the PK-DB API.

How data is curated

Curators extract measurements from publication tables and text, digitize figure panels, and record the study design and experimental conditions. Numerical data is stored in spreadsheets; structured metadata links those data to the subjects and interventions. Shared terms and units make records comparable while source references and curator comments preserve their provenance.

PK-DB overview: publication and raw data become curated measurements and metadata, which are validated and uploaded for exploration and reuse

Overview from the PK-DB documentation. Curation connects source data to validated records for browsing, analysis and modeling.

Curators check their entries against the source publication and resolve validation errors before upload. PK-DB also validates uploaded data. Automated checks support this process; scientific review establishes whether an entry faithfully represents the experiment.

Curating with this repository

pkdb_data holds the study source files in studies/, organized by substance and study. Each study folder keeps its metadata, spreadsheets and publication material together so that changes can be reviewed and tracked in Git.

  • Getting started: set up VS Code and the curation tools, configure access, synchronize terms, check and upload studies.
  • Curation guidelines: follow the curation workflow and learn how to organize study files and encode the experimental data.