Table 5 The side-by-side analysis of the sequencing variables considered in this study provides a comprehensive evaluation of their respective strengths and limitations
Variables | Advantages | Disadvantages |
|---|---|---|
PCR cycles | ||
Low number | • Reduced PCR amplification time • More accurate reads | • Loss of identification of rare species in the sample |
High number | • Allows detection of bacteria in low-biomass samples, such as skin or urine | • Requires longer time • Introduction of misclassified reads |
Annealing Temperature | ||
48°C | • Sequencing results more accurate than 52°C temperature with LongAmp polymerase | |
52°C | • Minimal differences in accuracy with iTaq polymerase | • Reduced amplification of Gram-negative bacteria with LongAmp polymerase |
Primer sets | ||
#1 (27F-1492R) | • Universally recognized primers for 16S PCR amplification • Minimal differences in accuracy with set#2 when used with LongAmp polymerase | • With iTaq polymerase, sequencing accuracy lower than set#2 |
#2 (GM3-GM4) | • With iTaq polymerase, sequencing accuracy higher than set#1 • With LongAmp polymerase, minimal differences in accuracy compared to set#1 | |
Taq Polymerases | ||
LongAmp | • Higher processivity than iTaq • Sequencing results more accurate • Recommended by ONT protocols | |
iTaq | • Results are not affected by the annealing temperature (range 48–52°C) • Lower price | • Substantially favour amplification of Gram-negative bacteria • Greater proportion of misclassified bacteria compared to LongAmp polymerase |
Workflows | ||
BugSeq | • Allows for accurate bacterial identification at both genus and species levels | • Subject to payment • Increasing PCR cycles significantly enhanced the percentage of misclassified read |
EPI2ME-16S | • Allows for accurate bacterial identification at genus level | • Limited capacity of customization • Workflow can be used only by ONT customers |
EPI2ME-WIMP | • Unsuitable for 16S based bacterial identification • Highest percentage of misclassified reads • Limited capacity of customization • Workflow can be used only by ONT customers | |
Kraken2 | • Allows for accurate bacterial identification at genus levels • Free of use workflow | • Salmonella is under-represented |
16S Databases | ||
RefSeq | • Non-redundant database • NCBI-managed database compiled from GenBank sequences | |
Silva | • Small and large rRNA subunits database including 16S rRNA sequences from the European Nucleotide Archive | |
Accuracy | ||
Low (> 80%) | • Analysis of all reads, including rare bacterial species | • Inclusion in the analysis of possible misclassified reads (~ 3% of the total) |
High (> 95%) | • Reliable sequencing results, no misclassified reads | • Over 50% loss of reads • Loss of depth in the analysis |