Reading an agarose gel after PCR

Reading an agarose gel after PCR is a selection. You decide whether a band is usable for the next step, or whether the lane has to be repeated, redesigned or sequenced before anyone trusts it. Four facts make that selection: the lane order you actually loaded, a marker that covers the expected size, the size you designed, and the no-template control. A match on size is not sequence identity. How the reaction produces that DNA is covered in how polymerase chain reaction works. How the gel sorts fragments is covered in agarose gel electrophoresis for DNA.

The decision you are allowed to make

Endpoint PCR plus a gel can support "a product of about this length is present" or "it is not detectable on this gel." It cannot support "this is the gene," "this many copies were in the sample," or "this band is pure enough to clone" unless you add a further check. Select the band for the claim you are making. A screening gel and a gel you will cut are different standards. The screening gel can tolerate a faint, single band. The gel you will excise needs that band separated from neighbours by enough agarose that the slice does not take them with it.

Write the claim in the notebook before you look at the photograph. People promote a lane after they see a glow. The criteria below are the promotion rules.

Four criteria, in order

Lane order comes first because every later sentence assumes it. Write the order before you load, load in that order, and photograph so the marker and the labels can be read together. A swapped no-template lane and sample lane turns contamination into a false positive, or a real product into a false scare. If you cannot reconstruct the order from the sheet and the picture, the gel is not usable, however pretty the bands.

The marker comes second. Use a ladder whose fragments bracket the expected amplicon, and do not estimate outside it. A band above the top fragment is only "larger than that fragment." A band below the bottom fragment is only "smaller." For products of a few hundred to a few thousand base pairs, a wide-range ladder on about 1 percent agarose is a common planning choice. Products near 100 base pairs usually need a higher percentage and a low-range marker. Follow the gel method you are actually running. The percentage changes both resolution and how far a given fragment travels.

Expected size comes third. You should already know it from the primer coordinates on the reference sequence, including both primers. Compare the band to the nearest marker fragments, not to a memory of where "about 500" usually sits. Resolution is limited. On a low-percentage gel, fragments that differ by a few dozen base pairs in the multi-kilobase range can look like one band. If your true product and a plausible off-target differ by less than that, this gel cannot select between them. Change the percentage or the run length, or stop claiming a distinction you cannot see.

The no-template control comes fourth and can veto the other three. That lane has every reagent except the sample DNA. If a band appears there at the sample's size, the sample lane is not evidence of the sample. A short, fuzzy band low in the no-template lane is usually primer-dimer. It vetoes a sample band that co-migrates with it, and it does not veto a sample band that sits well above it and is absent from the control. No marker, no veto power: a control you cannot locate on the photograph is a control you did not run.

What else in the lane is allowed to influence you

Tracking dyes are not DNA. Their position depends on the agarose percentage, and a dye front can sit near a small amplicon or a primer-dimer. If the only "band" is exactly where the dye runs, and it is the colour of the loading dye in white light, do not call it a product. Stain and photograph under the conditions the stain requires, with the orange safety shield in place. The hero image shows that shield. Bare ultraviolet light is an eye, skin and DNA hazard. Ethidium bromide and alternative nucleic-acid stains both need the safety data for the bottle you opened.

An overloaded lane glows thick and can smile at the edges. Thickness makes a band look larger than a properly loaded neighbour. If size is the decision, rerun less volume rather than arguing with a swollen band. A smear through the lane is not a set of hidden specific products. You may still select a discrete band inside a mild background if, and only if, you can cut or score it without the background, and the no-template lane lacks it. If you cannot, the lane is not usable for that purpose.

A positive-control lane, when you have a trusted template, tells you the mix and the programme can make the right size. It does not identify the sample band. Sample and positive can share a size and be different sequences. Use the positive lane to avoid blaming every blank sample on the enzyme.

When to accept, repeat, or go to sequence

Accept for a presence-and-size screen when all four criteria pass and the band is discrete. Record the marker name, the percentage, and the photograph. Accept for excision only when neighbouring bands, including primer-dimer, are far enough away that the slice is one species. If they are close, change the gel before you change the biology.

Repeat the PCR when the no-template lane is guilty, when the marker failed to resolve, or when the order is uncertain. Repeating onto the same contaminated primer dilution repeats the artefact. Make a fresh working dilution from a clean stock.

Go to a sequence, or to a restriction cut whose site lies inside the amplicon, when the claim is identity. Choose the enzyme from the reference sequence in the NCBI Nucleotide database, and expect one predictable pattern. An uncut band of the right size does not satisfy that claim. Neither does a chromatogram you never compared to the map.

Reject the lane when the product sits outside the ladder, when the no-template control matches it, or when the gel was stopped so early that dimer and amplicon are still one glow. A rejected lane is a result. It is not a cue to increase the contrast until something appears.

Safety, heat and a run that stopped midway

The gel buffer and the stain are the chemical hazards. Ultraviolet is the physical one. Do not hold a gel over a bare box while you look for a camera. Photograph once, with the shield, and with the marker in the frame. Disposal of stain and of amplified DNA follows the institutional rule. This selection guide is not a diagnostic result and not a biosafety approval.

A warm room can soften a low-percentage gel before a long run finishes, and bands then drift. A power cut mid-run leaves fragments that have not reached the separation you planned. Do not score that gel as if it had finished. Restart the separation, or accept that sizes are unreadable, and keep the marker in the picture so the next person can see the run was short. Humidity that wets the photograph label is a smaller problem than a label written an hour later from memory. Write the lane order on the image the day you expose it.

What to ask for

If you are sourcing markers, agarose, stain or polymerase for this readout, name the amplicon length and the percentage you need so the marker brackets it. The molecular biology catalogue is the family map. Send the length and the readout with the quote request. A ladder that does not cover your size cannot rescue the call. Where the gel is the last step of a sample path, mention the nucleic acid analysis pathway. If several products must be scored in one lane, the multiplex PCR enquiry reference is the design question, and the gel percentage is part of that question rather than an afterthought.