Hi Eric Hawrylyshyn

I find your question interesting. There are several methods of data normalization: the Boxcox transformation, logarithm application, data centering and reduction, etc. Each has its advantages and disadvantages. No method is perfect.

The works of the authors below will answer your question in advance, but you'll need to know the R language.

Finding Optimal Normalizing Transformations via bestNormalize by Ryan A. Peterson,2021

The bestNormalize R package was designed to help users find a transformation that can effectively normalize a vector regardless of its actual distribution. Each of the many normalization techniques that have been developed has its own strengths and weaknesses, and deciding which to use until data are fully observed is difficult or impossible. This package facilitates choosing between a range of possible transformations and will automatically return the best one, i.e., the one that makes data look the most normal. To evaluate and compare the normalization efficacy across a suite of possible transformations, we developed a statistic based on a goodness of fit test divided by its degrees of freedom. Transformations can be seamlessly trained and applied to newly observed data and can be implemented in conjunction with caret and recipes for data preprocessing in machine learning workflows. Custom transformations and normalization statistics are supported.

Ref : https://journal.r-project.org/articles/RJ-2021-041/RJ-2021-041.pdf

Warm Regards,

Abdi-Basid ADAN

Abdi-Basid Adan , I think that OP means "normalize" in the sense of "adjust for a factor", not "make follow a normal distribution".

Eric Hawrylyshyn , I'm not really familiar with that analytical equipment. It may make sense to divide by the control as you suggest. Or it might make sense to subtract the control. I would imagine there's a common procedure for how to handle these controls.

The fact that you found no significant difference among IP controls may not be dispositive. You should probably adjust for the controls in whatever manner is usual.

BTW, in statistical analysis, you could simply include the average IP control value as a covariate for each plate.

But as I mentioned, I'm not familiar with this kind of chemical / biological analysis, so take everything I say with a grain of salt.

I understand what you Sal Mangiafico mean, but the point here is to give you an idea of the best approach to standardization, and that's the purpose of the question, if you don't mind me saying so. The type of standardization chosen will be the distribution procedure thereafter, but first identify.

Warm Regards,

Abdi-Basid ADAN