I'm working on comparing bacterial metabolic models. Each models consists of a set of metabolites.
e.g Model1 - {M1, M2, M3, M4} (M1- metabolite name)
Model2 - {M3, M4, M5, M6, M7, M8}
I have to compare the difference between these 2 models, generate a dissimilarity matrix and finally cluster them. One suggestion I received on computing the dissimilarity matrix is "compute pair-wise Euclidean distances for each of the metabolites and sum up these differences". I'm bit confused on how to compare an Euclidean distance in this case as these are not numbers.
I'm looking for a theoretical explanation on how to achieve this.