Site-directed mutagenesis primers

Worked example

• A 25-base template with base 13 changed from A to C, in the overlapping complementary scheme, at 250 nM oligo in 50 mM Na⁺ with the primer in excess: 735 candidate windows considered, 734 off this short linear template, 1 passing every rule.
• The candidate: forward ACGTACGTACGTCTGCATGCATGCA (25 nt, 52.0 percent GC, 12 matched bases 5′ of the change and 12 3′ of it, Tm 63.97 °C on the mutant) with its exact reverse complement TGCATGCATGCAGACGTACGTACGT. The Tm of each primer's matched 3′ segment against the template is 43.68 °C for the forward primer and 40.48 °C for the reverse.
• The melting temperature of a primer against a template it does not fully match is outside this model and is not given; the two numbers above are both of perfectly matched duplexes. This is the published fixture case hand-substitution-overlap, worked by hand in the derivation script before the engine was written.
• A sequence design under the rules stated here. The melting temperature of a primer against a template it does not fully match is outside this model and is not given. No amplification, mutagenesis efficiency or colony outcome is predicted, and no protocol, polymerase or kit is recommended.

Method

A window of the mutant carrying the change with its reverse complement (overlapping), or the change at the forward primer's 5′ end with a matched partner abutting it (back-to-back); the matched flanks counted from the outermost changed base, with each primer's Tm against the mutant it matches and its matched 3′ segment's Tm against the template

• Site-directed mutagenesis primers. Two schemes, each a sequence design under the rules stated here. Overlapping complementary primers: a window of the mutant top strand containing the change with its exact reverse complement, annealing length 25 to 45 bases measured on the mutant, at least 10 perfectly matched bases on each side of the change (12 to 15 preferred), the change central to within 3 bases, GC 40 to 60 per cent preferred and 30 to 70 per cent required. Back-to-back primers, circular templates only: the changed bases within the 5′-most six bases of the forward primer with at least 15 perfectly matched bases 3′ of the last change, 18 to 30 perfectly matched bases on each primer, each primer's GC fraction 30 to 70 per cent as in the overlapping scheme, the partner primer's 5′ end abutting the forward primer's, and the pair's matched-part melting temperatures within 3 °C; the product is the linear whole-template amplicon and requires circularisation before it is a plasmid. Substitutions of 1 to 3 bases, insertions and deletions of up to 12 bases, templates of 2 to 50,000 bases. Melting temperatures come from the suite's nearest-neighbour model at the conditions shown, and every one of them is of a perfectly matched duplex: for each primer, its melting temperature against the mutant sequence it matches exactly, and the melting temperature of its longest perfectly matched 3′ segment against the template. The melting temperature of a primer against a template it does not fully match is outside this model and is not given. Candidates are ranked by flank balance, the preferred flank band, GC distance and length, each term compared at a published precision: that precision is the agreement between this engine and its oracle, not a physical resolution, and a melting-temperature difference of 1e-6 °C is not a difference anyone can measure. The counts of candidates rejected under each rule are reported. Nothing is relaxed to produce a result.Checked against 10 design cases over the two schemes and 8 refusals published and hashed from an independent derivation written before the engine (every melting temperature and reverse complement from Biopython 1.88 under the recorded conditions; the mutant, the candidate windows, the matched flanks, GC, the ranking and the counts by reason derived in the script; one case per scheme worked by hand with a write refusal): every sequence, coordinate, flank count, count and refusal exact, each melting temperature within 1e-6 °C (17 October 2026). The cases include a three-base substitution with an unchanged middle base, a twelve-base insertion, a twelve-base deletion crossing a circular origin, a short linear template whose only window carries unequal flanks, and a GC-rich stretch where no window passes: 405 of its 735 candidates fall to the matched-flank minimum, 242 to the central tolerance, 73 to the GC range and 15 to the template's ends.
• Oligo melting temperature by nearest neighbour. SantaLucia J Jr. A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proc Natl Acad Sci USA 1998;95:1460-1465 (Table 2; read 6 October 2026); concentration models after SantaLucia J Jr, Hicks D. Annu Rev Biophys Biomol Struct 2004;33:415-440.Checked against Biopython 1.88 MeltingTemp.Tm_NN within 0.1 °C on 45 oligos under both concentration models, with the Owczarzy 2008 regime exact and the seven refusals exact (6 October 2026).
• Monovalent salt correction. Owczarzy R, You Y, Moreira BG, et al. Effects of sodium ions on DNA duplex oligomers: improved predictions of melting temperatures. Biochemistry 2004;43:3537-3554 (equation 22).Checked against Biopython 1.88 MeltingTemp.Tm_NN within 0.1 °C on 45 oligos under both concentration models, with the Owczarzy 2008 regime exact and the seven refusals exact (6 October 2026).