DATAHOWLAB New Release

To assess process stability and generally understand intrinsic process variability, analysing the replicate error is a very important step.

Replicate Mapping is a new DataHowLab feature that will enable the analysis of replicate errors. Replicate mappings allow users to create groups of experiments that were executed following the same experimental recipe.

Within the groups, the relevant replicate error metrics are calculated automatically. These give insights into the process stability and the intrinsic process variability. Replicate groups can be compared to each other for every parameter.

Both the process variables and the CQAs can be considered for the replicate error metrics, in a flexible intuitive way.

Every project starts with an initial design, whether it’s process development, process characterization, scale-up or anything else.

This new feature allows to create space-filling designs based on Latin-Hypercube-Sampling, as an alternative to the traditional DoE designs. These experimental designs are optimal for working with hybrid models, and allow the models to capture the complex non-linearities of bioprocesses.

The initial designs can be created as part of an existing project or at the start of a new project. All relevant variables can be chosen manually, or easily selected from a reference project.

With this new design optimization algorithm, we’re simplifying and elevating this feature.

A new special algorithm for design optimziation was implemented. It is producing new recipes, while considering the model uncertainty and target values for the objective. Simple choices for either space exploration, exploitation or a balance between the two make the choices simpler.

Have a look at the Webinar by our CEO Dr. Alessandro Butté for a more detailed presentation of the strengths and advantages of this approach.