Numerical interpretation of TRS-PCR profiling results for Escherichia coli strains isolated from patients with bacteriuria in Lodz region, Poland.

With the multiplicity of existing methods to track E. coli infections, it still seems necessary to seek new, better and/or complementary ways for epidemiological investigations. Particularly, fast, cheap, effective and reproducible methods providing easily comparable results are needed. Our previous studies showed that the use of TRS-PCR is an effective molecular tool in E. coli epidemiology. In this paper, we have developed a unique classification scheme in which an individual TRS-PCR pattern is assigned a numerical value. This approach allows for rapid interpretation of the results obtained from several similarity dendrograms. Using this approach, based on CAC-PCR, GTG-PCR and CGG-PCR, we obtained 52, 86 and 99 different numerical types for the 124 analyzed uropathogenic E. coli strains, respectively. This allowed for the identification of 121 unique isolates differing in at least one TRS-PCR class. In this approach, we got numerical results, easy to sort and interpret, allowing easier analysis of these strains.