%20(1)-1.png?width=469&height=469&name=Black%20and%20White%20Modern%20Illustrative%20Support%20Service%20Logo%20(1%20x%201%20in)%20(1)-1.png)
Metrum Research Group — Training Programs
World-class pharmacometrics education from the team advancing model-informed drug development
Efficient Reproducible Bayesian Population PK with NONMEM
A hands-on workshop covering efficient, reproducible, traceable Bayesian population PK analyses — with decision-grade outputs built for real drug development workflows.
Upcoming Trainings
Tiered Pricing Available. Academic/Regulatory and Student pricing is available. Please review pricing details before completing enrollment.
Course Calendar
| Course | Date(s) | Time (US ET) | Status | Action |
|---|---|---|---|---|
| Simulation Concepts with mrgsolve | Nov 19–21, 2025 | 10:00 AM – 2:00 PM | Completed | Closed |
| Generalized Linear Models in R | Apr 7, 9, 14, 16, 2026 | 8:00 AM – 12:00 PM | Completed | Closed |
| Efficient Reproducible Bayesian Population PK ★ | May 19–21, 2026 (Tue–Thu) | 10:00 AM – 2:00 PM | Open | Register |
| Causal Concepts for Planning Analyses | Jun 10–11, 2026 | 10:00 AM – 2:00 PM | Coming Soon | Register |
What Past Attendees Say
Past Trainings
🔔 Be First to Know
New courses are added throughout the year. Drop your email and we'll notify you when the next training is announced — no spam, unsubscribe any time.
Detailed Course Information
All courses include: slides, example code/data, and an execution-ready Metworx instance per attendee with full download rights at course close.
A 1.5-day hands-on workshop covering efficient, reproducible, traceable Bayesian population PK analyses using NONMEM within the MeRGE open-source ecosystem.
Prerequisites
- Working NLME/pop-PK knowledge, practical NONMEM experience, R fluency
- Intro Bayesian concepts (Bayes' rule, priors; no prior MCMC required)
Learning Objectives
- End-to-end Bayesian PPK workflow: priors → sampling → checks → decisions
- Implement METHOD=BAYES/NUTS in NONMEM; fit PPK models with Bayesian methods
- Diagnose MCMC: chains, ESS, R̂; resolve convergence issues
- Model comparison with PPC/VPC and LOO; reproducible audit-ready pipeline
A DAG-first workshop giving pharmacometricians the conceptual understanding, vocabulary, and analytic tools to plan analyses with causal clarity using dagitty and R.
Prerequisites
- Comfort with simple regression and exposure-response modeling; no prior causal inference required
- Format: 1-day (2 × 3–4 hr sessions); mini-lectures + exercises in dagitty and R
Learning Objectives
- Formulate clear causal questions and estimands for PMx analyses
- Sketch and critique DAGs for PK/PD, adherence, selection mechanisms
- Identify minimally sufficient adjustment sets; distinguish confounding vs. mediation
- Map DAG identification to methods: outcome modeling, IPW/propensity scores
This workshop equipped pharmacometricians to build, fit, and diagnose exposure–response models using a pragmatic blend of maximum likelihood and Bayesian approaches in R — from GLM/GAM workflows to brms, with decision-grade outputs.
Course Information
- Format: 4 × 4-hour live remote blocks; mix of lecture, discussion, worked examples, and hands-on coding exercises
- Uses public-domain R packages only
- No local R installation required — each student received a web-based RStudio interface on Metworx with all packages pre-installed
Prerequisites
- RStudio/Posit fluency
- Familiarity with fitting linear models in R would be helpful
- Previous exposure to Bayes theorem would be helpful
Learning Objectives
- Specify and fit GLM models by MLE in R; compare and interpret those models
- Implement Bayesian GLMs using the brms package; run and interpret MCMC
- Diagnose models via residuals, visual predictive checks (VPCs), and posterior predictive checks
- Select priors for PK/PD contexts; troubleshoot sampling pathologies
- Produce decision-grade outputs: covariate forest plots, E-R summaries
This 1.5-day workshop taught pharmacometricians to turn PK/PD models into robust simulations using mrgsolve — covering population/batch simulation, VPCs, parallelized scenarios, and PTS calculations.
Topics Covered
- Load and validate mrgsolve models; construct event objects; control simulation horizons
- Scale from individual to population and batch workflows
- Visual Predictive Checks (VPC/PPC), parallelized scenarios, PTS calculations
- Build a Simulation Map™ linking questions → scenarios → metrics → go/no-go thresholds
Ready to Register?
Secure your seat in our upcoming courses — spaces are limited.