CPDSE · Data Science Implementation Dashboard

Filter:

Courses × Competency Categories

Advanced (A)

Conditional (C)

Basic (B)

Not covered

A Advanced

C Conditional

B Basic

— Not covered

DS Implemented

In Progress

Planned

Not Started

Categories from the CPDSE Pharmaceutical Data Science Competency Model (2024). Competency mapping is indicative — validate with course coordinators.

Project Goals

By H1 2029, all core pharmacy bachelor courses at SDU and UCPH integrate data science competencies at Conditional level or above — ensuring graduates leave with a coherent, progressive data science skill set aligned with the CPDSE Pharmaceutical Data Science Competency Model.

Data Science Learning Progression · BSc

	Year 1: Data, Noise & Scientific Questions	Year 2: Complexity, Confounding & Multivariable Models	Year 3: Prediction, Validation & Decisions
Theme	From observation to explanation	Understanding systems, not single effects	From models to consequences
Intent	Data as measured reality, not truth Variation as a fundamental property of biological systems Simple models as tools for thinking, not prediction engines	Most pharma questions are multivariable Context and confounding matter Models encode assumptions	Prediction is different from explanation Model performance is not enough Decisions require uncertainty awareness
Conceptual focus	Where pharmaceutical data comes from Noise, replicates, and uncertainty Visualisation as a scientific tool Linear relationships and residuals Formulating answerable scientific questions	Multivariable experimental design Confounding and batch effects Non-linear and dose–response relationships Sensitivity to modelling choices Ethical interpretation of data	Prediction vs inference Model validation and generalisation Uncertainty quantification ML and AI in pharmaceutical contexts Communicating results for decision-making
Learning outcome	Students can formulate a simple pharmaceutical question, explore data visually, fit a linear model, and explain observed variation and uncertainty.	Students can model multivariable pharmaceutical data and reason about how assumptions and confounders affect conclusions.	Students can evaluate predictive models, communicate uncertainty, and judge whether a model is appropriate for pharmaceutical decision-making.

Implementation Timeline · 2026 – 2030

Task

Both universities

SDU

UCPH

Studyboard deadline (placeholder)

Timeline is indicative. Studyboard deadlines are placeholders — confirm dates with studyboards. PDS = Pharmaceutical Data Science.