Hello,
I am currently working with microarray data and am unsure about the proper order of operations for preprocessing, specifically whether transformation or normalization should be performed first.
In its initial format, my data is structured with samples as columns and genes as rows. However, for disease classification, I need to transpose the data such that samples become rows and genes become columns to match the features properly.
After initially applying Combat normalization to my data (still in the samples-as-columns, genes-as-rows configuration) and visualizing it via a boxplot, the results appeared fine.
However, following data transposition (to a samples-as-rows, genes-as-columns configuration) and visualizing with a boxplot, the data looks markedly different and somewhat odd.
Could you please provide some guidance regarding this issue? Specifically, I am interested in understanding the appropriate order for data transformation and normalization in the context of microarray data preprocessing.
Thank you!
