Skip to content

Positioning

What OpenPKFlow is, what it is not, and how to place it next to commercial and regulatory systems.

What OpenPKFlow is

OpenPKFlow is a transparent, reproducible, open-source Python workflow for:

  • Dissolution similarity (f1, f2, bootstrap f2, MSD, model fitting, multi-media, SUPAC-IR screening helpers, alcohol dose-dumping f2 assessment)
  • Noncompartmental analysis (NCA), including steady-state and urinary workflows, plus model-informed sparse oral screening
  • Analytical PK simulation (1- and 2-compartment models, transit absorption, SS metrics)
  • IVIVC Level A (Wagner-Nelson, Loo-Riegelman, convolution, Levy plot, %PE) and Level B/C helpers (MDT / MRT correlation)
  • Multi-stage study pipeline orchestration (openpkflow.pipeline / study run)
  • Bioequivalence screening (2x2 TOST, power / sample size, research-grade replicate)
  • Pharmacometric reporting (Markdown, HTML, PDF, DOCX)
  • Research-grade population PK diagnostics and estimation helpers (estimation frozen)

The product focus is a formulation-to-report pipeline: load study data, run transparent calculations, and export shareable reports for formulation, PK, and regulatory review teams.

Philosophy:

A transparent, reproducible, open-source Python workflow for dissolution, NCA, PK/PD simulation, and pharmacometric reporting.

What OpenPKFlow is not

OpenPKFlow is not:

  • FDA-approved software
  • A validated system under 21 CFR Part 11 or ICH Q2 by itself
  • A replacement for expert regulatory judgement
  • A full substitute for commercial platforms used under controlled, locked, regulated environments (e.g. Phoenix WinNonlin, NONMEM production pipelines)
  • EMA ABEL, full-replicate RSABE, or any formal replicate-design BE method without independent validation fixtures. FDA partial-replicate RSABE remains NOT_EVALUABLE until its external validation gate is complete.
  • A drop-in Part 11 validated quality system without site SOPs and controls
  • Full SUPAC automation (v2.6 helpers are screening only)
  • A primary regulatory sparse-PK method; the one-compartment oral sparse fit is a model-informed screening tool and requires study-specific validation

Final regulatory interpretation must always be reviewed by qualified formulation, pharmacokinetic, and regulatory experts.

Current release context (for agents)

Version 2.6.0 was published to PyPI and GitHub Releases on 2026-07-15. Always read HANDOFF.md at the repository root in the working tree for branch name, PR number, verification results, and release status.

Where formal BE and PopPK live

Capability OpenPKFlow role Formal / companion path
2x2 crossover TOST Convenience layer for screening and teaching Study SOPs + independent statistical review
Power and sample size Exact TOST power (PowerTOST-aligned) Confirm with PowerTOST or study statistician
RSABE / replicate designs Partial screening only Companion BioEqPy package (out of scope here)
Population PK FOCE-I / SAEM Research-grade; limited model space Pharmpy, nlmixr2, NONMEM for primary NLME

Population PK estimation in OpenPKFlow is frozen for extension beyond bug fixes. Treat PopPK results as research and method-development tools, not as a primary regulatory NLME engine.

Formulation-to-report pipeline

Typical sequential workflow (library APIs or CLI):

  1. Dissolution -- f1/f2 (and optional bootstrap) on multi-batch profiles
  2. NCA -- per-subject AUClast, Cmax, half-life, CL/F (or CL for IV)
  3. BE -- paired TOST on NCA-derived parameters (or pre-computed AUCs)
  4. Report -- HTML / Markdown / PDF / DOCX with the mandatory disclaimer

IVIVC Level A sits beside this path for extended-release development when dissolution and IV / oral PK support a correlation.

See Pipeline tutorial for CLI and Python walkthroughs, and examples/pipeline_walkthrough.py for a dependency-light script.

Validation transparency

Cross-checks against public and published references are documented in:

Executable tests under tests/validation/ lock those claims. Agreement with a comparator does not make OpenPKFlow a Part 11 validated system. Regulated use still requires local SOPs, version control, locked environments, and independent review.

Claims to avoid

Do not describe OpenPKFlow as:

  • "FDA-approved"
  • "replaces Certara" / "replaces Phoenix"
  • "Part 11 validated out of the box"
  • "regulator-grade RSABE"
  • "AI discovers the perfect formulation"

Use transparent, reproducible, open-source language instead.