Smears and extra bands in PCR

A smear or a handful of extra bands means the tube made DNA of more than one length. Sometimes the length you designed is still in there as a discrete, cuttable band. Sometimes it is not, and another cycle will only add glow. This page is about that choice: when to keep endpoint PCR and cut the band, and when to stop and change the primers, the template or the programme. The reaction that filled the lane is described in how polymerase chain reaction works. Separating what you cut is agarose gel electrophoresis for DNA.

When a messy lane is still the right tool

Use the lane when you can point to one band at the expected size, the marker brackets it, and the no-template control is empty at that height. Background above or below is a warning about purity, not an automatic failure. If you only need a presence-and-size call, score the discrete band and record the background. If you need DNA to clone or sequence, cut only when the band is separated from the smear and from any neighbour by enough gel that the blade takes one species. A slice that includes the smear becomes the smear in the next reaction.

Stop using the lane when there is no discrete band, when the expected size is fused to another band, or when the no-template lane shows the same pattern. In those cases the gel has already answered. More exposure, more cycles, or a longer look at the ultraviolet box will not create an edge that was not synthesised. Redesign, or change the condition that made the extra DNA, and run a new tube.

Endpoint brightness is still not a copy number. A smear can hide a successful amplicon or a failed one. The shape of the lane is the evidence, not how strongly the camera saturated.

Where the extra DNA comes from

Too many cycles is the common self-inflicted smear. Early cycles make the intended product. Later cycles, after primers or nucleotides tighten, copy mismatched ends, truncated strands and rare off-target starts. Those products differ by small lengths and run as a smear, often with the original band still visible inside it. The experiment that tests this is a repeat at a lower cycle count, not a new enzyme. If the lower count restores a single band, keep that count. If the band is already single at 25 cycles and smears only at 40, you do not have a primer emergency.

Too much template does two different things, and you should tell them apart. A lane can be overloaded: the band is real and so thick that it grins at the edges and looks high-molecular-weight. Dilute the load on a new gel. The same reaction, loaded at a fraction, should sharpen. Too much template inside the PCR itself also promotes mis-priming, because weak primer matches find DNA more often, and it can even inhibit the enzyme so you get a weak smear instead of a band. The test is a fresh reaction with less template, often a ten-fold dilution as a comparison, not a new gel of the same tube. Follow the amount the enzyme card expects.

A low annealing temperature lets partial matches survive long enough to extend. You see extra bands at discrete wrong sizes more often than a featureless smear. Raise the temperature in a short series. The extras should fade before the intended band does. If they fade together, the intended product was never much more stable than the junk, and the primers need a new 3-prime end.

High magnesium stabilises duplexes, including the weak ones. It is a specificity problem wearing a chemistry label. Extra bands that appear only after someone added magnesium, and that shrink when magnesium returns to the mix's stated level, have been explained. A common classical planning range is about 1.5 to 2.5 millimolar, and the mix you opened may already contain it. A second addition is not a rescue. Treat any magnesium change as its own experiment, with a no-template lane, because dimer loves the same condition.

Degraded template gives a smear because the polymerase extends from broken molecules and falls off at ends that are already there. Genomic DNA that was sheared on a warm bench, or RNA that was later reverse-transcribed from fragments, produces a ladder of lengths with or without a weak specific band. Loading less does not turn that into one size. A careful re-extraction does. If a plasmid prep is smeared below the supercoiled band before you even amplify, fix the prep. PCR will amplify the damage.

A short path before you redesign

Change one cause at a time, or you will not know which habit was guilty. First, load less of the existing product so overload is not mistaken for chemistry. Second, if the product was cycled past the point where the band was already visible, repeat with fewer cycles. Third, if extras are discrete, raise the annealing temperature. Fourth, if magnesium was increased, return it to the card. Fifth, if the template was old, warm, or already smeared, prepare it again.

Redesign when that path leaves extra bands stuck to the product, or when the no-template lane grows the same short band at every temperature you try. Move the 3-prime ends. A pair that dimers will keep dimerising while you titrate everything else. Redesign is the stop rule, not the first reflex and not a failure of nerve.

Homologous genes produce extra bands that no magnesium setting will delete, because both sites are real. If a second locus of known size appears, either accept a gel that can resolve the two lengths or move a primer onto a base that differs. Cycling harder copies both.

Safety and templates that sat out

The gel is the hazard in the readout: stain chemistry and ultraviolet light. Use the shield and the safety data for the stain you poured. A smeared amplicon from a regulated sample is still that sample's containment problem. This page is not a diagnostic interpretation.

Smears have a climate. Template left in a warm extraction tray breaks, and the PCR then reports the breakage as a smear. If the first gel was faint because the run was short or the load was small, adding ten cycles on a new attempt is how a recoverable band becomes a smear by the afternoon. Record the cycle count you add. In humid weather, a gel that sits uncovered before you photograph it can diffuse at the edges. Score from a prompt picture with the marker in frame, not from a gel that has been sweating on the bench.

What to say when you ask for reagents

Say whether the lane is a smear, a thick overload, or discrete extras, and what you already changed: cycles, template amount, annealing temperature, magnesium. Name the amplicon length. Enzyme families are in the molecular biology catalogue. Send the lane description with the quote request. A difficult extract that arrives already degraded belongs with the nucleic acid analysis pathway, because a new polymerase will copy the fragments you give it. Several products in one tube that have become an unreadable ladder are a design question for the multiplex PCR enquiry reference, not a reason to keep the same primer set and add cycles.