1. Multilocus Variable-Number Tandem-Repeat Analysis of Enterohemorrhagic Escherichia coli Serogroups O157, O26, and O111 Based on a De Novo Look-Up Table Constructed by Regression Analysis
- Author
-
Kentaro Kawatsu, Hidemasa Izumiya, Yusuke Takahashi, Tetsuya Harada, Takao Kawai, Yuki Wakabayashi, and Nao Umekawa
- Subjects
0303 health sciences ,040301 veterinary sciences ,030306 microbiology ,Prediction interval ,Regression analysis ,04 agricultural and veterinary sciences ,Computational biology ,Multiple Loci VNTR Analysis ,Biology ,bacterial infections and mycoses ,medicine.disease_cause ,Applied Microbiology and Biotechnology ,Microbiology ,0403 veterinary science ,03 medical and health sciences ,Molecular typing ,Variable number tandem repeat ,Lookup table ,medicine ,Animal Science and Zoology ,Escherichia coli ,Food Science - Abstract
Multilocus variable-number tandem-repeat analysis (MLVA) is a widely accepted molecular typing tool for enterohemorrhagic Escherichia coli (EHEC). However, ensuring the accuracy of MLVA data among multiple laboratories remains difficult. We developed a method of constructing adjusted look-up tables, which are necessary for MLVA profiling, at each laboratory using a regression analysis based on electrophoresis data from 24 in-house reference strains. On performing MLVA against 51 EHEC O157 isolates, the repeat numbers of 46 isolates were determined accurately using the look-up table with a 99% prediction interval, an outcome superior to that when using a 95% prediction interval. For the remaining five isolates, although the electrophoresis size fell outside the look-up table, we were able to predict the repeat number accurately by extrapolation or the nearest values of the look-up table. Our approach provides more accurate results than a nonadjusted conventional look-up table for calibrating MLVA profiles.
- Published
- 2021
- Full Text
- View/download PDF