Primer Dimer Checker

Paste a primer, or a forward and reverse pair, to check it for hairpins, self-dimers and cross-dimers. Each structure comes with its ΔG, its base pairing, and whether it pairs a 3′ end, which decides whether a dimer can be extended.

Runs in your browser. Sequences are not uploaded.

Primer 15′ → 3′

Primer 2 (optional, for a pair)5′ → 3′

Reaction conditions25 °C · 50 mM Na+ · 1.5 mM Mg2+

ΔG temperature°CNa+ / K+mMMg2+mMdNTP (total)mMPrimer, for TmnM

Defaults: Primer3 salt and primer concentrations, ΔG at 25 °C.

Examples

No problem structures

Every hairpin and dimer found is weaker than the tolerated ΔG limits.

Strongest structure found by each check
Check	Strongest ΔG	Result
Primer 1 hairpins	None	OK
Primer 1 self-dimers	None	OK
Primer 2 hairpins	None	OK
Primer 2 self-dimers	−0.45 kcal/mol	OK
Cross-dimers	−0.45 kcal/mol	OK

Primer length, GC content and melting temperature
Primer	Length	GC	Tm
Primer 1	19 nt	57.9%	57.2 °C
Primer 2	20 nt	45.0%	53.7 °C

Tm difference 3.5 °C. Nearest-neighbor Tm (SantaLucia 1998) at 50 nM primer.

OK: ΔG above -6 kcal/mol for dimers and -3 for hairpins, stricter at the 3′ end. Strong: below -9, or hairpin Tm of 47 °C or more.

Primer 1 hairpins

No hairpin with ΔG below 0 at 25 °C.

Primer 1 self-dimers

No dimer with ΔG below 0 at 25 °C.

Primer 2 hairpins

No hairpin with ΔG below 0 at 25 °C.

Primer 2 self-dimers

OKΔG −0.45 kcal/mol3 bp, includes a 3′ end

5′ GAAGATGGTGATGGGATTTC 3′
   |||
3′ CTTTAGGGTAGTGGTAGAAG 5′

Cross-dimers

OKΔG −0.45 kcal/mol3 bp, includes a 3′ end

5′ GAAGGTGAAGGTCGGAGTC 3′
   |||
3′ CTTTAGGGTAGTGGTAGAAG 5′

OKΔG −0.45 kcal/mol3 bp, includes a 3′ end

5′ GAAGGTGAAGGTCGGAGTC 3′
         |||
3′       CTTTAGGGTAGTGGTAGAAG 5′

How the primer dimer check works

The checker looks for three kinds of secondary structure and reports how stable each one is as a Gibbs free energy. The more negative ΔG is, the more of the primer is tied up in that structure.

Hairpins. Part of a primer pairs with another part of the same primer and folds it back on itself. The checker looks for stems of 3 or more base pairs around a loop of 3 or more bases.
Self-dimers. Two copies of the same primer pair with each other. Every alignment of the primer against itself is scored.
Cross-dimers. The forward primer pairs with the reverse primer. Checked when you enter both primers of a pair.

ΔG = ΔH − T × ΔS

1Every stretch of 3 or more consecutive Watson–Crick pairs is found.
2ΔH and ΔS are summed over each pair of neighboring base pairs (the nearest-neighbor stacks), plus an initiation term for each end of the stretch, or a loop penalty for a hairpin.
3The salt correction adjusts ΔS for the Na+, Mg2+ and dNTP concentrations.
4ΔG is calculated at the chosen temperature, 25 °C by default, and compared with the limits in the next section.

Parameters and conditions

Stacking and initiation values are SantaLucia's 1998 unified nearest-neighbor parameters. Hairpin loop penalties and the terminal A·T penalty are from SantaLucia and Hicks (2004).

Salt changes ΔS by 0.368 × (N − 1) × ln[Na+]eq for a stretch of N base pairs. Mg2+ counts as sodium through [Na+]eq = [Na+] + 120 × √([Mg2+] − [dNTP]), in mM, because dNTPs bind part of the Mg2+.

The default conditions are Primer3's: 50 mM Na+, 1.5 mM Mg2+, 0.6 mM dNTP and 50 nM primer. Primer Tm uses the same nearest-neighbor parameters and salt correction as Primer3, so it can be compared with Primer3 and Primer-BLAST at those settings.

Only consecutive Watson–Crick pairs are scored. Mismatches, bulges, dangling ends and G·T pairs are not, so a structure held together by two stretches on either side of a mismatch is shown as its stronger stretch. Use the checker to screen primers, and look closely at anything flagged Check.

What ΔG is too strong

A structure is tolerated when its ΔG is above the limit for its type. Limits are stricter when a 3′ end is paired, because that is the end the polymerase extends.

ΔG limits in kcal/mol by structure type
Structure	OK	Check	Strong
Hairpin, 3′ end in the stem	−2 or above	below −2	below −9, or Tm ≥ 47 °C
Hairpin, 3′ end free	−3 or above	below −3	below −9, or Tm ≥ 47 °C
Dimer that pairs a 3′ end	−5 or above	below −5	below −9
Dimer away from the 3′ ends	−6 or above	below −6	below −9

ΔG in kcal/mol.

The tolerated values follow PREMIER Biosoft's primer design guidelines. The −9 kcal/mol line is from IDT's primer design guidance, and 47 °C is Primer3's default maximum hairpin Tm. They are rules of thumb, not hard cutoffs.

The 3′ end matters most. A primer whose 3′ end is paired with another primer can be extended into a short primer-dimer product in every cycle. Complementarity near the 5′ end, such as a restriction site added as a tail, leaves the 3′ ends free and is much less likely to cause trouble.

For hairpins, Tm is often easier to read than ΔG. A hairpin that melts well below your annealing temperature opens before the primer needs to bind.

What a primer dimer is

A primer dimer is a short product that forms when two primers anneal to each other through complementary bases and the polymerase extends them. Because it is short, it amplifies efficiently and competes with the real product for primers, dNTPs and polymerase.

On an agarose gel a primer dimer shows up as a faint or fuzzy band below about 100 bp. In SYBR Green qPCR it adds signal to no-template controls and a second melt-curve peak at a lower temperature than the product.

How to fix primer dimers and hairpins

Change the 3′ end. Move the primer a few bases along the template, or change its last bases, so they are not complementary to the primer itself or to its partner. Keep one or two G or C among the last five bases.
Use a hot-start polymerase. Dimers form most easily while reactions sit at room temperature during setup. A hot-start enzyme stays inactive until the first denaturation step.
Raise the annealing temperature. Try a few degrees higher, or use touchdown PCR, so short complementary stretches melt before they are extended.
Lower the primer concentration. Fewer primer molecules means fewer primer–primer encounters. Many protocols work at 0.1 to 0.5 µM of each primer.
Set up reactions on ice. Add the polymerase last and move the tubes straight into a preheated cycler.

Worked examples

Both primers are in the checker's examples, at the default conditions, so you can check every number.

A self-dimer at the 3′ end

ACCTGAACAGATCCGCATGC ends in GCATGC, which is its own reverse complement, so two copies of the primer pair over their last 6 bases.

1Stacks GC, CA, AT, TG and GC: ΔH = −43.8 kcal/mol, ΔS = −114.6 cal/(K·mol).
2Initiation at two G·C ends (ΔH +0.2, ΔS −5.6) and the correction for a self-complementary duplex (ΔS −1.4): ΔH = −43.6, ΔS = −121.6.
3Salt: [Na+]eq = 50 + 120 × √(1.5 − 0.6) = 163.8 mM, so ΔS changes by 0.368 × 5 × ln 0.1638 = −3.33, giving ΔS = −124.93.
4ΔG at 25 °C = −43.6 − 298.15 × (−0.12493) = −6.35 kcal/mol.

−6.35 kcal/mol with a 3′ end paired, past the −5 limit: Check. Changing the last base to A (…GCATGA) breaks the palindrome, and the strongest self-dimer drops to −1.70 kcal/mol.

A hairpin

In GCGCTTCGGAAGCGCATGTACC, bases 1–6 (GCGCTT) pair with bases 10–15 (AAGCGC) around a 3-base loop (CGG).

1Stem stacks GC, CG, GC, CT and TT: ΔH = −45.9 kcal/mol, ΔS = −119.2 cal/(K·mol).
2Loop of 3 bases: ΔG37 = +3.5 kcal/mol, all entropic, so ΔS = −3.5 / 310.15 × 1000 = −11.29.
3Salt on the 5 stem stacks: ΔS −3.33, giving ΔS = −133.81.
4ΔG at 25 °C = −45.9 + 298.15 × 0.13381 = −6.00 kcal/mol. Tm = −45,900 / −133.81 − 273.15 = 69.9 °C.

The hairpin stays folded up to about 70 °C, above typical annealing temperatures and above the 47 °C limit: Strong.

Related primer and sequence tools

Primer dimer FAQ

What is a primer dimer?

A short PCR product made when two primers anneal to each other instead of to the template and the polymerase extends them. It happens when primers are complementary, especially at their 3′ ends. Primer dimers use up primers and polymerase, lower the yield of the real product, and give false signal in SYBR Green qPCR.

How do I check if my primers will form dimers?

Paste the forward and reverse primer above. The checker aligns each primer with itself and with its partner, scores every complementary stretch with nearest-neighbor ΔG, and flags structures past the usual limits, with a stricter limit when a 3′ end is paired.

What ΔG is acceptable for a primer dimer?

As a rule of thumb, dimers weaker (less negative) than −6 kcal/mol are tolerated, or −5 kcal/mol when a 3′ end is paired, and anything stronger than −9 kcal/mol is likely to cause problems. For hairpins the tolerated values are −3 kcal/mol, or −2 kcal/mol with the 3′ end in the stem.

What is a primer hairpin, and what does hairpin Tm mean?

A hairpin forms when one part of a primer pairs with another part of the same primer, folding it back around a loop. Hairpin Tm is the temperature at which half of the molecules are folded. Above it the primer is mostly unfolded and free to bind the template, so a hairpin with a Tm well below your annealing temperature usually does little harm.

What is the difference between a self-dimer and a cross-dimer?

A self-dimer, or homodimer, forms between two copies of the same primer. A cross-dimer, or heterodimer, forms between the forward and the reverse primer. Enter both primers to check all three.

Why does the 3′ end matter?

DNA polymerase only extends a primer whose 3′ end is paired. A dimer that pairs a 3′ end can be copied into a primer-dimer product. Complementarity that leaves both 3′ ends free cannot be extended, so the limits for it are less strict.

What does a primer dimer look like on a gel?

A band below about 100 bp, often faint or fuzzy, running well ahead of the product. It is strongest in reactions with little template and in no-template controls. In SYBR Green qPCR it shows up as a second melt-curve peak at a lower temperature than the product.

How do I get rid of primer dimers?

Redesign the primer so its 3′ end is not complementary to itself or its partner, use a hot-start polymerase, raise the annealing temperature, lower the primer concentration, and set up reactions on ice.

Why do other tools give different ΔG values?

They use different parameters, conditions and structure models. This checker scores consecutive Watson–Crick pairs with SantaLucia nearest-neighbor values at the conditions shown. Folding programs such as UNAFold also score mismatches, bulges and dangling ends, which usually makes ΔG a little more negative. Compare values within one tool and at the same conditions.

Is my sequence uploaded anywhere?

No. Every calculation runs in your browser.

References

SantaLucia J. A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. PNAS 1998;95:1460–1465.
SantaLucia J, Hicks D. The thermodynamics of DNA structural motifs. Annu Rev Biophys Biomol Struct 2004;33:415–440.
von Ahsen N, Wittwer CT, Schütz E. Oligonucleotide melting temperatures under PCR conditions. Clin Chem 2001;47:1956–1961.
Untergasser A, et al. Primer3 — new capabilities and interfaces. Nucleic Acids Res 2012;40:e115.