Pathogenicity and detection of virulence genes are necessary for completing the identification criteria for an organism. So PCR test for on of the virulence genes or pathogenicity using experimental animals are two main cases enable us to know wheather the organism is pathogenic or no.
I agree With Yagmur that you will probably need a combination of looking at serotypes and virulence genes and the easiest was it to use molecular Methods. Are there any specific E. coli pathogens you are thinking about? For classical STEC O157:H7 you do have selective plates where you are able to distinguish these from nonpathogenic E. coli, but the others usually don't have any biochemical marker.
As Camilla and Yagmur said serotyping can help you very much because there are some serotypes that are considered pathogenic this paper could help you Lior, H., 1994. Classification of Escherichia coli. In: Gyles, C.L.
Ž . Ed. , Escherichia coli in Domestic Animals and Humans.
Cab International, UK,
And also you have to do PCR´s for finding specific virulence factors you are looking for in your strains. Toxins, pilus, capsule, Types of secretion systems are among many others are important for virulence
did you distinguish successfully between pathogenic and non pathogenic strains?
if I decided to perform PCR, which genes should i search for, as encoding for virulence factor?
as for the pathogenicity test on experimental animals, how should I introduce the tested bacteria to the animal, and what are the signs or symptoms supposed to appear if the tested bacteria (isolated from urinary tract of human) is pathogenic?