Nearest-neighbor Tm (SantaLucia 1998)
This is the accurate method, the one primer design software uses. It sums the stacking energy of every adjacent base pair instead of just counting bases, then corrects for how much oligo and salt you actually have. Use this number for ordering primers; use the rules above for coursework.
Secondary structure screen
A primer that folds back on itself or sticks to a second copy of itself will not prime well. This is a base-pairing screen, not a folding free-energy prediction, so treat it as a flag to investigate rather than a verdict.
How to calculate primer Tm and GC content
The melting temperature is the temperature at which half the DNA strands separate, critical for setting PCR annealing temperatures. For short primers (14 bases or fewer) the simple Wallace rule works well: Tm = 2 × (A + T) + 4 × (G + C), because G-C pairs have three hydrogen bonds and A-T pairs only two. For longer sequences use Tm = 64.9 + 41 × (G + C − 16.4) / N, where N is the length. GC content is just (G + C) / N × 100. The reverse complement is what you get by pairing each base (A↔T, G↔C) and reading the new strand 5′→3′, the sequence of the opposite strand, and the basis for designing a reverse primer.
Method note: the two rules above are the standard teaching formulas, and they are salt-independent and concentration-independent. The nearest-neighbor calculator on this page does the accurate version, accounting for base stacking, oligo concentration, and salt. For a typical 16 to 20 base primer the two agree within a few degrees, but for short oligos the Wallace rule runs far too hot: ATGCAT comes out at 16 °C by the rule and about −11 °C by nearest-neighbor at 250 nM. Use the nearest-neighbor number when a PCR has to work.
Looking for OligoCalc? The Northwestern link now 404s. Here is where it moved, and how it compares to this page.
Related tools: DNA → protein translation · Gene expression regulation explorer · Codon chart (genetic code) · all biochem tools.
Worked example 1: the default 20-mer primer
GCGGTCCTAAGCTTGGATCC, 20 nucleotides, so this uses the long-sequence formula (not Wallace, which only applies at ≤14 nt).
This matches the calculator's default output exactly, 55.9°C, 60.0% GC, and that reverse complement.
Worked example 2: a short palindromic primer (Wallace rule)
ATGCAT, only 6 nucleotides, so this uses the simple Wallace rule instead.
ATGCAT is a palindromic sequence. Its reverse complement is itself. This isn't a coincidence: it's the actual recognition sequence for the restriction enzyme NsiI. Restriction enzymes almost always recognize palindromic sequences like this, because the enzyme is a symmetric dimer that binds both strands identically.
FAQ
Why two different Tm formulas?
Wallace (2×(A+T)+4×(G+C)) works for short oligos (≤14 nt) but overestimates Tm for longer sequences. The GC-content formula (64.9+41×(G+C−16.4)/N) is calibrated for longer sequences instead. See the contrast in the two examples above.
Why does higher GC content raise Tm?
A G-C pair forms three hydrogen bonds; an A-T pair forms only two. More bonds means more energy needed to separate the strands, so GC-rich sequences melt at a higher temperature than AT-rich ones of the same length.
What is the reverse complement used for?
DNA is double-stranded and antiparallel, so the opposite strand is the complement read in reverse. You need it to design a primer binding the opposite strand (a reverse PCR primer), or to know what the other strand reads at a given region.
Why might real PCR software give a different Tm?
These are salt- and concentration-independent teaching formulas. Real tools like Primer3 use nearest-neighbor thermodynamics accounting for adjacent base pairs, salt, and primer concentration, typically shifting Tm by a few degrees from these simple formulas.
Practice problems
1. Find the Tm and GC% of the short primer GCATCG.
Show answer
2. Find the Tm and GC% of ATGCGATCGTAGCTAGCTAGGCTA (24 nt).
Show answer
Sources and how to cite this page
For short oligonucleotides this uses the Wallace rule, Tm = 2(A+T) + 4(G+C), from Wallace, R.B. et al., Nucleic Acids Research 6(11):3543–3557, 1979. For longer sequences it uses the GC-content formula with a salt correction, which accounts for the fact that Tm rises with both GC content and cation concentration. GC pairs contribute more because they form three hydrogen bonds to AT's two, and because of stronger base stacking.
The nearest-neighbor calculation. This page implements the unified nearest-neighbor parameters of SantaLucia, J., “A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics,” PNAS 95(4):1460–1465, 1998. It sums the ΔH and ΔS of each adjacent base-pair step, adds the helix initiation terms, and applies the −1.4 cal/(mol·K) symmetry correction for self-complementary sequences. Salt is handled the way primer design software does it: the monovalent correction of Owczarzy, R. et al., “Effects of sodium ions on DNA duplex oligomers,” Biochemistry 43(12):3537–3554, 2004, and the magnesium correction of Owczarzy, R. et al., “Predicting stability of DNA duplexes in solutions containing magnesium and monovalent cations,” Biochemistry 47(19):5336–5353, 2008, selected automatically from the ratio √[Mg2+] / [Na+]. Because dNTPs chelate magnesium, the free Mg2+ is solved for first rather than using the total. Two independent checks: the thermodynamics reproduces the published worked example CGTTGA/TCAACG (ΔH° −41.2 kcal/mol, ΔS° −115.4 cal/(mol·K), ΔG°37 −5.4 kcal/mol), and the complete pipeline agrees with Biopython's Tm_NN to 0.0 °C across sodium-only, magnesium, dNTP, and self-complementary test cases.