Normalization Overview:

The normalization procedures are grouped into three categories. You can use one or combine them to achieve better results.

• Sample normalization is for general-purpose adjustment for systematic differences among samples;
• Data transformation applies a mathematical transformation on individual values themselves. A simple mathematical approach is used to deal with negative values in log and square root Please search OmicsForum using "normalization #metaboanalyst" to find more information.
• Data scaling adjusts each variable/feature by a scaling factor computed based on the dispersion of the variable.