DNA primer melting temperature (Tm)

SantaLucia (1998), ΔH and ΔS are summed stack by stack (ten independent parameters), with one initiation term per end and a correction when the duplex is self-complementary (deduced from the sequence, never asked of the user). C_eff equals the total primer concentration for a self-complementary duplex, or a quarter of it for two distinct strands at equal concentration. The salt correction only affects entropy, over a calibration range of roughly 0.05 to 1.1 M Na⁺. An ambiguous IUPAC base makes this calculation fail rather than invent a value, see the simple estimate below. Which does bound that uncertainty.

The Wallace rule (Marmur & Doty, 1962) applies to short oligonucleotides and directly counts A/T and G/C bases. For longer sequences, the GC-adjusted formula (also due to Marmur & Doty) uses the number of G/C bases and the total length N. Neither accounts for primer or salt concentration: always shown alongside the thermodynamic model above for comparison, never as a substitute.

Tm

· melting temperature: the temperature at which 50% of double-stranded molecules are denatured (single-stranded).

ΔH, ΔS

· total enthalpy and entropy of the duplex, negative for base pairing (an exothermic process), in kcal/mol and cal/(mol·K) respectively.

Effective concentration (C_eff)

· the concentration actually used in the Tm equation, which depends on whether the duplex is self-complementary (see above).

Self-complementary

· a strand equal to its own reverse complement, which therefore pairs with a copy of itself, distinct from a plain textual palindrome.

GC content

· proportion of guanine and cytosine bases in the sequence, the higher it is. The higher the Tm (3 hydrogen bonds versus 2 for A-T).

N

· total sequence length, in number of bases.

Wallace rule

· historical empirical rule for short oligos (classic PCR primers), fast but approximate.

• Only monovalent cations (Na⁺/K⁺) are accounted for; magnesium and dNTPs are not corrected for (Owczarzy et al., 2008, not adopted for lack of independent verification at the same standard).
• Both strands are assumed to be at equal concentration; the unequal-concentration variant is not implemented.

The thermodynamic model remains incomplete: it only accounts for monovalent cations (Na⁺/K⁺), not magnesium or dNTPs. Whose usual correction (Owczarzy et al., 2008) was not adopted for lack of independent verification at the same standard. It also assumes two strands at equal concentration, the unequal-concentration variant is not implemented. An ambiguous IUPAC base makes it fail rather than produce a range: the simple estimate below then bounds the Tm instead. That estimate remains an empirical approximation based only on GC content and length, ignoring the exact base order, primer and salt concentration, and any mismatches. Order of magnitude of the gap between the two methods, measured on a panel of reference primers (M13, T7, T3, SP6, gapdh, actb, 16S): roughly 1–2 °C typical on a standard 18–30 base primer, up to 5 °C in the least favorable cases, and considerably more at composition extremes, where the simple estimate (Wallace rule) diverges by over 15 °C for a 100% A/T sequence and 13 °C for a 100% G/C one. This gap is not an error in either method: the two rest on different assumptions.

• 0.05 ≤ sodium ≤ 1.1model · warning

  The salt correction is calibrated for roughly 0.05 to 1.1 M Na⁺ (SantaLucia, 1998); outside this range, extrapolation is not guaranteed.

  Instead · Bring the salt concentration closer to the calibrated range, or treat the result as indicative only.

• strand ≥ 0physical · refusal

  Primer concentration cannot be negative or zero.

  Instead · Check the entered concentration value and unit.

For the short primer atgcatgcat (10 bases, 4 G/C), the simple estimate gives Tm = 2×6 + 4×4 = 28 °C. For a longer 20-base primer at 50% GC: Tm = 64.9 + 41×(10−16.4)/20 ≈ 51.8 °C. The thermodynamic model, by contrast, depends on the exact base order and the entered ionic conditions: two primers of the same length and GC content can have different nearest-neighbor Tm values.

1. SantaLucia, J. · Proceedings of the National Academy of Sciences (PNAS), A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics, PNAS 95(4):1460–1465 (1998), DOI 10.1073/pnas.95.4.1460.

   Location : Table 1 (paramètres ΔH°/ΔS° plus-proche-voisin et termes d’initiation), pages 1460–1463 · accessed 2026-08-01 · DOI 10.1073/pnas.95.4.1460

Scope and limitations

• The comparison estimate (Wallace rule / GC-adjusted formula) is a historical heuristic; its exact attribution by primer length has not been verified to the same standard as the nearest-neighbor model above.

Provisional · Sources and assumptions are declared; the full documentary review is still pending.

Physical model. The result follows the significant figures of the least precise input.

revision 1 · reviewed 2026-08-01

Revision log

1. revision 1 · 2026-08-01 · source update

   Bloc scientifique renseigné : référence SantaLucia 1998 formalisée, hypothèses et domaine de validité déclarés.

   Aucun changement de résultat, les paramètres plus-proche-voisin ne sont pas modifiés.