AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

pkpd-modeling

Pharmacokinetic and pharmacodynamic modelling and simulation - non-compartmental analysis, compartmental and population PK, PK/PD and exposure-response, TMDD, PBPK orientation, bioequivalence, allometric scaling and first-in-human dose, drug interaction prediction, and Bayesian therapeutic drug monitoring. Use when analysing concentration-time data, deriving exposure metrics, fitting PK or PD models, or evaluating dosing regimens. Triggers include "pharmacokinetics", "pharmacodynamics", "PK/PD", "NCA", "non-compartmental", "AUC", "Cmax", "lambda z", "half-life", "clearance", "volume of distribution", "compartmental model", "population PK", "popPK", "NONMEM", "nlmixr2", "Pharmpy", "Monolix", "exposure-response", "Emax", "EC50", "indirect response", "effect compartment", "TMDD", "PBPK", "bioequivalence", "RSABE", "ABEL", "allometric scaling", "first-in-human", "MABEL", "drug-drug interaction", "DDI", "ICH M12", "concentration-QTc", "therapeutic drug monitoring", "MIPD", and "dosing regimen".

Original source
K-Dense-AI
License
MIT
Pinned revision
5f2bc0341354fd5590aef76a533ada1e1e5b4e62

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read k-dense-ai--pkpd-modeling at revision 5f2bc0341354fd5590aef76a533ada1e1e5b4e62, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 17

references/source-ledger.md

Source ledger — reviewed 2026-10-01

Execution evidence

The bundled CLI suite and synthetic smoke examples were run in isolated Python with NumPy 2.5.3 and SciPy 1.18.1. Tests cover analytical linear-PK identities, NCA, individual fitting, ordinary BE, DDI arithmetic, TDM, scaling, event validation and PD helpers. Analytical and synthetic agreement is numerical verification, not clinical validation or qualification for a regulatory application.

Pharmpy 2.2.0 was installed in a separate isolated environment. Seven transformations on the bundled pheno model executed: QSS TMDD, indirect/direct/effect-compartment PD, allometry, transit absorption and Michaelis-Menten elimination. Signatures for those methods, annotate_unit and calculate_eta_shrinkage were inspected. Nineteen run_* tools were enumerated. No licensed NLME estimation, R workflow or PBPK simulation was executed.

Current interfaces and package metadata

Scientific/regulatory corrections and primary evidence

  • ICH M12 full text: TDI5×Cmax,u, distinct MATE/systemic efflux cutoffs, fu reliability, inlet/enterocyte concentrations and separate mechanism evaluation. Per-minute/per-hour consistency and full-inhibition ceilings were checked algebraically.
  • FDA BE May 2026: design-specific replicate models, squared-contrast bias correction and upper chi-square quantile. The CLI now rejects unsupported raw replicate/scaled analysis. PowerTOST informs planning comparisons; the bundled quantile-grid method remains an approximation.
  • Phoenix8.4 guide: automatic terminal selection differs from the helper's strict extension-improvement rule.
  • Vancomycin consensus: serious MRSA, total AUC24 400–600 mg·h/L with MIC 1 mg/L assumption; not a universal antimicrobial index target.
  • E11A, E14/S7B, FIH, M13A/B/C and M15 primary sources and regional dates are listed in regulatory guidance. M13B is now EMA Step 5, not draft Step 2b.

Deliberate boundaries

No script calls a remote clinical/data API; it reads local CSV/CLI values. External URLs are primary documentation links, not service/authentication endpoints. Phantom endpoints were not introduced. No patient data, outcome validation, licensed engine run, prospective clinical workflow or submission review was performed. Product-specific guidance is not enumerated. Sparse-data identifiability, priors, censoring assumptions, parameter uncertainty and extrapolation remain the analyst's responsibility; the supplied tools cannot resolve them from convergence alone.