Question repeatedly asked in various forms:

"Dear developer, I would like to use NormalyzerDE for a dataset containing less than 15 datapoints. Is this possible, and what are the limitations?"

Dear user,

NormalyzerDE is designed for normally distributed (i.e. proteomics, metabolomics, microarray) omics-datasets which normally contains hundreds- or thousands- of datapoints. Some of the normalizations and many of the performance measures will not be reliable for tiny datasets.

With this said, you can obtain 'partial run' results also for smaller datasets. This includes output from the standard normalization techniques and a subset of the performance measures in the evaluation report. Note that you need to be careful with interpretation here, and base the choices of normalization techniques in a good understanding of the data rather than only using the report.

This reduced report is currently provided by the web interface when running a dataset with less than 50 datapoints. If running in R you can adjust the threshold using the parameter "tinyRunThres", which by default has the value 50.

Best wishes, Jakob