troubleshooting
Multiplex PCR primer compatibility
How to tell whether multiplex primer pairs share a window, avoid cross-dimers, and make sizes a gel can separate when the long product drops out.
- Author
- EVRINTH Editorial Team
- Published
- 8 October 2026
- Updated
- 8 October 2026
- Reading time
- 7 min

Multiplex PCR puts several primer pairs in one tube and asks each pair to make its own product. Compatibility means three things at once: the pairs share an annealing window, they do not extend one another as cross-dimers, and the products land at sizes your gel can tell apart. A pair that is flawless alone can still drop out in that company. The long product is usually the first to go. This is a troubleshooting path for that failure. The single-reaction cycle is described in how polymerase chain reaction works. Endpoint readout is still presence and size, not a count of starting copies.
What the mix is competing for
Each cycle, every primer that has annealed can take polymerase and dNTPs. Short products reach the opposite primer sooner, become full templates, and multiply. A long product that is only half finished is a poor template for the next cycle. If the extension hold was copied from the shortest assay, the long amplicon never joins the exponential phase. That is why "it worked in singleplex" and "it vanished in the mix" can both be true. The singleplex extension was long enough. The shared programme is not.
Primers also compete at the sequence level. In singleplex you only inspect two oligos for 3-prime complementarity. In a six-primer mix you inspect every oligo against every other oligo. A cross-dimer uses the 3-prime end of a primer from pair A and a primer from pair B. It needs no template, it is short, and it appears in a no-template control that contains the full primer set. It will not appear in the singleplex no-template controls. Those controls were the wrong experiment.
All of this happens at one annealing temperature. If one pair only works several degrees below the others, the compromise set-point either drops that pair or dirties the rest. Calculated melting temperatures are a start. The shared window is the temperature where every intended size appears and the full-set no-template lane stays free of those sizes and of a new short band.
Reagent classes that change the balance
The enzyme is whichever polymerase the card allows for multiplex, often a hot-start class so the extra primers do not get extended during the longer setup. Nucleotides and magnesium stay at the mix's stated levels unless you are running a deliberate one-variable test. Adding magnesium because two bands look weak changes every pair at once.
Primer concentration is the lever people reach for, and it is easy to overdo. Each primer is often kept toward the lower end of the usual endpoint band, around 0.1 to 0.5 micromolar, because six oligos at the top of that band are a lot of 3-prime ends. Follow the mix. Lowering only the pair that is winning, while leaving the others fixed, is a readable experiment. Lowering everything at once is not.
The gel is part of compatibility, not a downstream courtesy. Choose a percentage and a marker that resolve the set, as in agarose gel electrophoresis for DNA. If two amplicons differ by less than the gel can show, redesign a length. No amount of enzyme will pull apart bands that co-migrate.
A path from symptom to cause
Start with the singleplex reactions at the annealing temperature you intend to share, not at each pair's private favourite. A pair that only works at its own temperature is not compatible yet. Include a no-template control for each pair, then one no-template control with every primer present and no template.
Add pairs in stages. Two pairs, then a third. The run where a band disappears names the primer you just added as a suspect for competition or a cross-dimer. Jumping straight to the full set only tells you that something is wrong.
If the longest band is the one missing, lengthen extension to what that polymerase needs for that length and repeat the staged mix before you redesign. Classical Taq-like planning is often about one minute per kilobase near 72 Celsius. Engineered enzymes publish other speeds. Follow the card. If the long band returns, the programme was the constraint. If it stays missing while short bands thrive, reduce the primer concentration of the strongest short pair as the next single change, or accept that those targets belong in two tubes.
If a new short band appears only in the full primer set, including the no-template lane, treat it as a cross-dimer. Find the 3-prime match between pairs and replace one oligo. More cycles will copy the short band more efficiently than the biology you wanted.
If every product is weak, including those that were strong alone, suspect the total primer load, a forgotten reagent, or inhibition in a template that the singleplex diluted differently. Do not add ten cycles as the first move.
If two bands sit so close that you cannot score them, the chemistry may be fine and the assay is still unusable. Change a product length or the gel percentage. Score only gaps you can see beside a marker.
| What you see | Likely constraint | What to change next |
|---|---|---|
| Long band missing, short bands strong, singleplex long band fine | Extension or competition | Time the extension for the long product, then ease the dominant pair's primers |
| New short band only when all primers are in, including the no-template tube | Cross-dimer | Replace the oligo whose 3-prime end matches another pair |
| One short pair missing at the shared temperature | That pair is outside the window | Redesign its melting temperature rather than lowering Ta for everyone |
| All bands weak only in the full mix | Primer load, enzyme demand, or a pipetting miss | Restore a staged mix before adding cycles |
| Bands present but not separable | Size design or gel percentage | Move an amplicon or change the gel, then rescore |
| Full-length band in the all-primer no-template lane | Carry-over of an old product | Discard working dilutions and clean the setup |
Failure modes that survive a careful mix
Homologous targets can prime across each other and produce a band that matches none of the designed lengths. If the loci share sequence, multiplex is the wrong convenience. Split them or design primers on the bases that differ, then check those primers in NCBI Primer-BLAST against both loci.
A positive control plasmid that carries only one target will light only one band. That does not prove the other pairs are compatible. Use a control that contains every target, or accept that each singleplex positive only clears its own pair.
Inhibition looks like a total dropout and is not fixed by primer redesign. Dilute the template. If the singleplex reactions were run with a diluted sample and the multiplex was run with a crude one, you changed two variables and learned nothing about compatibility.
Very different starting amounts make the abundant target eat the reaction. Multiplex is a poor way to see a rare sequence beside a dominant one. Move the rare target to its own tube rather than cycling the mix into a smear.
Safety and the specification you write
Each extra primer is another oligonucleotide, not a new biosafety category. The template still is. Follow the institutional decision for that sample, and treat amplified product as a contamination source for the next setup. This page is research troubleshooting. It is not a validated diagnostic panel.
When you ask another group to build the assay, write the targets, the size gaps you can resolve, the enzyme class, and the cycler. A request that says only multiplex PCR leaves the window and the gel unstated, which is how a received primer set fails on the bench that ordered it. In a shared instrument room, record the ramp and the block. A programme timed for a fast cycler can starve the long product on a slow one even though the extension line looks identical.
What to put in the enquiry
Name the number of targets, their approximate lengths, whether each pair already works alone, and which band drops out. Enzyme and nucleotide families are listed in the molecular biology catalogue. Send that description with the quote request. The multiplex PCR enquiry reference is a prompt for this design conversation. It does not mean an assay is already running or that a particular panel exists. If the template is an extract with a difficult matrix, say so through the nucleic acid analysis pathway so inhibition is not mistaken for incompatible primers.
Questions from the bench
Why does a pair that works alone disappear in the multiplex?
Alone, that pair has the polymerase, the nucleotides and the annealing temperature to itself. In the mix it competes with every other pair, and it may form a cross-dimer that did not exist in the single reaction. A clean singleplex result is a prerequisite, not a prediction.
Which product should I expect to lose first?
The longest one. Short amplicons finish sooner, copy more often, and spend dNTPs and enzyme that the long product still needs. An extension time set for the short products leaves the long one incomplete. Give the programme the longest product, then look again.
How far apart do the band sizes need to be?
Far enough that your gel percentage and run length put daylight between them next to a marker. A gap that looks generous on a map can close up for large fragments on a low-percentage gel. If two products sit on top of each other, the multiplex cannot be scored from that gel no matter how well it amplified.
Can I fix a dropout by adding more cycles?
Usually that brightens whichever product is already winning, including a cross-dimer. Find which pair fails in staged mixes, check cross-complementarity, and set extension from the long product. Add cycles only after the missing band is faintly present and the no-template lane is still clean.
References
Manufacturer names identify published method classes. Trademarks remain with their owners. Catalogue records on this site are independent references for enquiry. They are not a statement of inventory, distribution rights or a supply commitment. This page is educational. It is not medical advice, a diagnostic protocol or a biosafety approval.
Catalogue
Related products and categories
These links follow the subject of the article into published manufacturer references. A listing is a reference for an enquiry, not a statement of stock or distribution rights.
Continue in this cluster
Related reading
How polymerase chain reaction worksHow PCR copies a chosen DNA segment: the temperature cycle, reagents, controls and what a band on a gel does and does not prove.
A glossary of PCR termsPlan a PCR by defining amplicon, Tm, Ta, hot-start, plateau, master mix, NTC, inhibition, fidelity and multiplex before you mix.
Choosing an annealing temperatureWhy annealing temperature is not the Tm printed on a tube, and how salt, magnesium, additives, and a short gradient locate the real window.