How is Fidelity Measured?

Return to High-Fidelity PCR

Fidelity comparisons between polymerases can be expressed in absolute terms, often by the number of errors per 1,000 or 10,000 nucleotides, or expressed as the number of theoretical errors per base. The level of fidelity can also be expressed in relative terms, often using Taq DNA polymerase as the relative standard. Note that while single base substitution errors are the easiest to assess, polymerases are also capable of insertion/ deletion errors that can lead to frameshift mutations, a highly deleterious type of error.

Comparison of High-Fidelity Polymerases

(Reported by supplier)
(For simple templates5)
(For complex templates5)
Q5 High-Fidelity
DNA Polymerase (NEB)
>100X Taq1,2 20 kb simple;
10 kb complex
10 s/kb 10 s/kb (<1 kb)
20–30 s/kb (>1 kb)
Phusion® High-Fidelity
DNA Polymerase (NEB)
>50X Taq1,2 20 kb simple;
10 kb complex
15 s/kb 30 s/kb
AccuPrime™ Pfx (Life) 26X Taq1 12 kb4 60 s/kb4
PfuUltra™ II Fusion HS
20X Taq1 19 kb4
15 s/kb (<10 kb4)
30 s/kb (>10 kb4)
PfuUltra High-Fidelity
DNA Polymerase (Agilent)
19X Taq1 17 kb simple;
6 kb complex
60 s/kb (<10 kb)
120 s/kb (>10 kb)
60 s/kb (<6 kb)
120 s/kb (>6 kb)
Platinum® Taq HiFi (Life) 6X Taq1 20 kb4 60 s/kb4
KOD DNA Polymerase
4X Taq3 6 kb simple;
2 kb complex
10–20 s/kb 30–60 s/kb

1 PCR-based mutation screening in lacZ (NEB), lacI (Agilent) or rpsL (Life).
2 Due to the very low frequency of misincorporation events being measured, the error rate of high-fidelity enzymes like Q5 is difficult to measure in a statistically significant manner. Although measurements from assays done side-by-side with Taq yield a Q5 fidelity values of approximately 200X Taq, we report “>100X Taq” as a conservative value.
3 Takagi et al (1997) Appl. Env. Microbiol. 63, 4504-4510.
4 Template not specified.
5 Simple templates include plasmid, viral and E. coli genomic DNA. Complex templates include plant, human and other mammalian
genomic DNA.
6 Values provided by individual manufacturers.


Polymerase fidelity assays take many forms and have been used extensively for comparing high-fidelity polymerases. The pioneering work of Thomas Kunkel (1) utilized portions of the lacZα gene in M13 bacteriophage to correlate host bacterial colony color changes with errors in DNA synthesis, using a variety of prepared M13 substrates for assessing single base nucleotide substitution errors or frameshift mutations. William Thilly (2) championed denaturing gradient gel electrophoresis to distinguish between heteroduplexes formed when error-containing PCR amplicon strands reannealed in the presence of an excess of wild type (WT) complementary strands (lower melting temperature), and the corresponding WT homoduplex form (higher melting temperature).

Another approach, championed by Wayne Barnes (3), utilizes 16 cycles of PCR to copy the entire lacZ gene and portions of two drug resistance genes with subsequent ligation, cloning, transformation and blue:white colony color determination. Similar to the Kunkel assays, most errors in the lacZ-encoding β-galactosidase gene cause a loss-of-function of the ability to utilize the Xgal substrate on agar plates that would normally result in a blue colony. This assay is presently used at New England Biolabs for large scale, relatively quick determination of DNA fidelity, as the 3,874 bp amplicon affords a reasonable sequence space for the scoring of DNA polymerase errors. Sanger sequencing of individual cloned PCR products can also score fidelity of the thermophilic DNA polymerase employed in the PCR as long as the error rate of the polymerase used for sequencing is lower than that of the PCR polymerase.

  1. Kunkel, T.A. and Tindall, K.R. (1987) Biochemistry, 27, 6088–6013.
  2. Ling, L.L. et al. (1991) Genome Research, 1, 63–69.
  3. Kermekchiev, M.B., Tzekov, A and Barnes, W.M. (2003) Nucl. Acids Res. 31, 6139–6147.