Video-on-Demand Library

12 April 2022

Speaker: Tim Holland-Letz

In practical dose response trials, especially in a cell culture context, experimental designs to establish functional relationships between dose and effect of a substance are generally chosen based on simple rules of thumb. These designs are generally reasonably effective, but not optimal. In this webinar we give an overview over the statistical optimal design approach in this context. We show how optimal designs can be constructed, covering mainly all-purpose designs (D-optimality) but also more specific designs (c-optimality). We introduce a graphical representation. Next, we show how these concepts can be extended to trials aiming to estimate the interaction between two different substances given simultaneously. Finally, we present an R-Shiny application which allows construction of optimal designs in most of these contexts with a minimum of theoretical knowledge.

Part of Collection(s)

Topic(s)

Upcoming Events

Latest Jobs