Abstract
The efficacy and reproducibility of the Phene-Plate (PhP) system (Biosys Inova, Stockholm, Sweden) for biochemical fingerprinting of Salmonella typhimurium was investigated. Duplicate and replicate assays on 40 epidemiologically related and unrelated strains were performed in two batches of PhP-48 plates. The data were then subjected to various numerical analyses to determine the infra-subspecific relationships of the isolates, and PhP-types determined by reference to a predetermined identity level. Our results show that, (i) presently defined identity levels are too stringent and delineate PhP-types which are epidemiologically unjustified, (ii) tests currently recommended for PhP-typing S. typhimurium may be somewhat unstable and not satisfactory for fingerprinting purposes, (iii) caution must be exercised when comparing data from different batches of PhP-48 plates, and (iv) best results are obtained when infra-subspecific relationships are determined by visual assessment of the resulting dendrogram. These data indicate the discriminatory capacity of the Phene-Plate system to be significantly compromised by its reproducibility, although epidemiologically valid data can be obtained if results are carefully examined.
Original language | English |
---|---|
Journal | Journal of Microbiological Methods |
Volume | 26 |
Issue number | 1-2 |
Pages (from-to) | 185-195 |
ISSN | 0167-7012 |
DOIs | |
Publication status | Published - 1996 |
Keywords
- reproducibility
- numerical analysis
- phenotyping
- epidemiology
- Salmonella