Akermi, are you talking about cluster analysis as mathematical tool to study similarity between objects? If so, the key is to introduce a proper distance between objects. Suppose, objects can differ in several dimensions, some of which are discrete (0 or 1, if property is present) and other are continuous (weight, length). Distance between components is introduced first. The overall distance can be for example Euclidean (square root of the sum of squares of component distances) or other. Note that proper choice of distance will influence the final result. Once it is introduced, dendrogram can be constructed using special software.

thank you for your help, but my object are microorganisms that have different caractérestiques each other is I want to know the similariture between these organisms gather for group

example :

(Bacillus cereus is the microorganism is its characteristic are: Gram (+), citrate (+) presence of catalase etc. )

( Staphylococcus aureus, Gram (-) , citrate (+), absence of catalase )

Akermi, I do not know biology, but I know math. You can set a list of those discrete features. If  for some feature, like Gram, both have plus, the distance in this component is zero. If for one microb (A) feature Gram is plus, but for other (B) minus, then - distance is one. Then each (microb) element is a vector (of CHAR type, its elements are symbols). If we have 2 microbs, A (+, +, -, -,+) and B (+,-, -, -, -), then the distance components will be  a vector (0,1,0,0,1). Now you can build a distance D by choosing for example the sum of all sub-distances; here D=2.

You can also weight different components, if you think that difference in one feature is more important than in the other. For example, you think that Gram is twice more important than citrate. Then consider 3 microorganisms: A(+, +), B(+, -), C(-, +), where the first element is Gram. Then the distance d(A,B)=1, while d(A,C)=2. Clearly, d(B,C)=2+1=3.

thank you, sir, I will try to understand because I too do not know math, then I will apply this aproach on microbiology.

Use GC%