OpenPKFlow¶

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

What it does¶

OpenPKFlow gives formulation scientists, PK/PD researchers, and CRO/CDMO teams a clean Python workflow for:

Module	What it covers
`dissolution`	f1, f2, bootstrap f2, MSD, max deviation, model-dependent comparison, Weibull/Higuchi/KP/first-order/zero-order model fitting
`nca`	AUClast, AUCinf, Cmax, Tmax, lambda_z, t1/2, CL/F, Vz/F — three AUC methods, explicit BLQ, %AUCextrap flag, DN params, CDISC PP
`be`	2x2 crossover TOST (convenience layer), GMR + 90% CI, intra-subject CV, BioEqPy export
`sim`	1- and 2-compartment IV bolus/infusion/oral, repeated dosing, superposition
`pop`	GOF plots (4-panel), VPC with percentile bands, NONMEM-style dataset helpers
`report`	Markdown, HTML, PDF (ReportLab), Word (python-docx)
`ml`	Experimental torch MLP surrogate for 1-cmt oral profiles
`validation`	Utility functions for cross-checking against reference values

It does not replace expert regulatory judgement or validated commercial platforms. It makes routine analysis faster, cleaner, and more reproducible.

Install¶

pip install openpkflow

For PDF and Word reports:

pip install openpkflow[reports]

For ML surrogate (torch):

pip install openpkflow[ml]

Quick example¶

from openpkflow.dissolution import f1, f2

reference = [20.0, 40.0, 60.0, 80.0, 90.0]
test      = [21.0, 39.0, 61.0, 79.0, 88.0]

print(f"f1 = {f1(reference, test):.2f}")   # 1.33
print(f"f2 = {f2(reference, test):.2f}")   # 72.80

See the Tutorials section for complete worked examples.

Philosophy¶

OpenPKFlow is report-first: every analysis ends in a clean, shareable output — HTML, PDF, or Word — suitable for supervisors, clients, CROs, and regulatory teams. Calculation correctness is necessary but not sufficient.

This package is open-source. Final regulatory interpretation should be reviewed by qualified formulation, pharmacokinetic, and regulatory experts.