Competitive binding-based optical DNA mapping for fast identification of bacteria--multi-ligand transfer matrix theory and experimental applications on Escherichia coli.
Bottom Line: Our identification protocol introduces two theoretical constructs: a P-value for a best experiment-theory match and an information score threshold.The developed methods provide a novel optical mapping toolbox for identification of bacterial species and strains.The protocol does not require cultivation of bacteria or DNA amplification, which allows for ultra-fast identification of bacterial pathogens.
Affiliation: Department of Astronomy and Theoretical Physics, Lund University, Sölvegatan 14A, 223 62 Lund, Sweden.Show MeSH
Related in: MedlinePlus
Mentions: As a proof-of-principle, we demonstrate that theoretical barcodes from different E. coli strains are sufficiently different so that the CB assay can differentiate them. To that end, we determined P-values for the 12 DNA fragments from strain CCUG 10979 with IS values larger than 100, matched to the theoretical barcodes for nine different E. coli strains. The strains are listed by NCBI as reference genome sequenced E. coli strains. A comparison of CCUG 10979 and these eight strains was performed and the resulting total BLAST scores are shown in Table 1. The total score is the sum of scores of all aligned sequences. Figure 10 shows that the average P-value is significantly lower for DNA fragments from strain CCUG 10979 than for any other strain. Four strains are statistically well separated from the correct one, three are on the very limit to be statistically resolved while strain O157:H7 Sakai is hardest to exclude. However, using the average of all P-values could give a false picture, since it will be strongly affected by a single ‘outlier’ that increases the average dramatically. Furthermore, fragments that have a high P-value for both strains compared should not be taken into account at all when comparing them. More information can potentially be obtained by, as in Figure 9, instead comparing the P-values for individual fragments when fitted to two strains. Figure 11 shows such a comparison for CCUG 10979 and O157:H7 Sakai (similar plots for all strains can be found in Figure S9 in the Supplementary Information). A vast majority of the fragments fit better to strain CCUG 10979. The major ‘false positive’ has a high P-value also for strain O157:H7 Sakai (∼20%) and should, therefore, not be taken into account when comparing the two strains. Rather, when comparing fragments that have a P-value lower than 5% for at least one of the strains, only 2 out of 9 fragments fit better to strain O157:H7 Sakai. The assay used is thus able to resolve these two strains, provided the IS cut-off and P-value tools are used.
Affiliation: Department of Astronomy and Theoretical Physics, Lund University, Sölvegatan 14A, 223 62 Lund, Sweden.