Which clustering method is preferable to use in Heatmap?

Question

[elefth.pavlos]

Hi,

I am trying to draw a heatmap, using ComplexHeatmap package, of lipidomics data (values of blood concentration). I am not sure whether euclidean distance or pearson is more relevant for my plot. The method has to be data specific or there is some rule to choose a method based on? Prior to plot my data, I usually log10 transform the matrix and then scale the variables.

Thanks!