BLOG AND NEWS
Comparison of data science workflows for root cause analysis of bioprocesses
As in a large number of industries, the biotechnology and pharmaceutical industries often resort to an analysis of the main causes of deviations or incidents, in particular with the help of tools such as Ishikawa diagrams.
However, these analyses are often very imprecise because the variables identified are not very close to the reality of events in bioprocesses.
Two approaches are then currently used in the framework of the analysis of principal causes, the analysis of raw data but also the analysis based on variables.
Today, these two techniques are able to explain the variance observed within a bioprocess. The two most commonly used tools are least squares regression and Principal Component Analysis.
The article compares the strengths and weaknesses of the two methods and shows that using these two approaches in a complementary way allows to gain in analysis efficiency and to deepen the knowledge on your biotechnological processes.
At Absolute RxD, Principal Component Analysis and Least Squares Regression have been studied during our training.
We have mastered these two techniques which will allow us to identify the most important variables of your processes and explain the observed variance. Also, thanks to our experience with UpStream and DownStream processes, we will be able to generate new hypotheses within your company, which can then be experimentally tested by your Research and Development teams.