Every band on a gel is a population of molecules that share roughly the same size, separated from larger and smaller neighbors by the sieving action of the gel matrix. Reading a gel means comparing where your sample bands sit relative to a set of known-size standards (the “ladder”), judging how bright each band is, and watching for patterns that signal something went right or wrong with your experiment. The logic is the same whether you are running DNA, RNA, or protein, though each type of molecule has its own quirks worth knowing.
Why Smaller Molecules Travel Farther
Gel electrophoresis works because charged molecules migrate through a porous gel when you apply an electric field. DNA carries a uniform negative charge per unit length, so when it enters an agarose or polyacrylamide gel, the only thing that determines how far it travels is size: smaller fragments slip through the gel pores more easily and end up farther from the wells where the samples were loaded. The relationship between distance traveled and fragment size is not perfectly linear but follows a pattern where distance is inversely proportional to the log of molecular weight.1PubMed Central. Agarose gel electrophoresis for the separation of DNA fragments That log relationship is what makes a gel image look the way it does: bands near the top (closest to the wells) are spaced far apart, while bands near the bottom are crowded together. The ladder lane makes this obvious because the gaps between its rungs get progressively tighter as you move down.
Using the Ladder to Estimate Fragment Size
The ladder, sometimes called a molecular weight marker, is a mixture of DNA or protein fragments whose sizes are already known. You load it into one or more lanes alongside your samples. After the gel runs and you stain it, you can see the ladder bands at predictable positions. To figure out the size of an unknown band, you find which ladder bands bracket it, then estimate where it falls between them.
For a rough read, eyeballing is fine. If your band sits halfway between the 500 base-pair and the 1,000 base-pair ladder bands, it is somewhere around 700 bp. For more precision, you can plot distance traveled against the log of fragment size for each ladder band, draw a best-fit line, then read off where your unknown band falls on that line. Software tools automate this by selecting the nearest marker bands and interpolating, which can push the sizing error down to about half a percent.2Bioinformatics. A computer program for molecular weight determination of DNA fragments (HOWBIG)
A few practical tips make ladder reading easier. Always load the ladder in a lane at the edge of the gel and, if you have many samples, in a lane somewhere in the middle too. Gels do not always run perfectly uniformly across their width, so having a nearby reference helps. Also make sure you know which ladder you are using: a 100 bp ladder and a 1 kb ladder look different, and mixing them up will throw your size estimates off by an order of magnitude.
How Gel Concentration Changes What You See
The concentration of agarose or polyacrylamide in the gel acts like a dial controlling the pore size. A low-concentration agarose gel (say 0.5 to 0.8 percent) has large pores, which lets big DNA fragments separate nicely but does a poor job distinguishing small ones. A high-concentration gel (1.5 to 2 percent) has smaller pores that resolve small fragments well but may barely separate large ones. This tradeoff means you should pick your gel percentage based on the size range you care about.3PubMed Central. Agarose and Polyacrylamide Gel Electrophoresis Methods for Molecular Mass Analysis of 5–500 kDa Hyaluronan
At higher concentrations, bands from small molecules become sharper and more distinct because the tighter matrix limits how much the sample diffuses sideways and vertically during the run. But very large molecules can run into trouble: they may struggle to thread through the small pores and end up compressed near the top of the gel, poorly separated from one another. In extreme cases, the gel essentially acts like a wall for the largest species, severely underestimating their real size if you try to read them off the ladder curve.3PubMed Central. Agarose and Polyacrylamide Gel Electrophoresis Methods for Molecular Mass Analysis of 5–500 kDa Hyaluronan
Gradient gels, which start at a low concentration near the top and increase toward the bottom, offer a workaround. Both large and small molecules get resolved in a single run because each molecule migrates until it hits the pore size that slows it down. Commercially available gradient polyacrylamide gels (such as 4–20 percent) can cover a broad size range in one shot, which is useful when you are not sure how big your fragments are.3PubMed Central. Agarose and Polyacrylamide Gel Electrophoresis Methods for Molecular Mass Analysis of 5–500 kDa Hyaluronan
What Band Brightness Tells You
A brighter band generally means more material. Staining dyes bind to nucleic acids or proteins, and the more molecules packed into a band, the more dye binds and the stronger the signal. For DNA, the observed brightness of a band is proportional to both the amount of DNA and the size of the fragments, because a longer fragment binds more dye molecules per strand.4PubMed Central. Deconvolution of nucleic-acid length distributions: a gel electrophoresis analysis tool and applications This means that if you have equal numbers of a 3,000 bp fragment and a 500 bp fragment, the 3,000 bp band will look roughly six times brighter, simply because each molecule is six times longer. Ladder manufacturers account for this by loading more mass in some bands than others so the brightness looks roughly even.
If you are trying to quantify how much DNA or protein is in a band (densitometry), be aware that the relationship between brightness and amount is only linear over a limited range. Load too little and the band disappears into background noise; load too much and the signal saturates, so doubling the amount no longer doubles the brightness. For protein Western blots, this saturation problem is well documented and can produce misleading results when researchers treat every band’s brightness as directly proportional to protein abundance without verifying that they are working within the linear range. Artifacts like “ghosting,” where an overloaded band of one protein bleeds into a neighboring region, can further skew readings. Careful researchers run a dilution series of their sample to confirm that the signal increases proportionally before trusting any quantitative comparison.
For gel-based protein detection, techniques using intrinsic fluorescence imaging have demonstrated linear quantification ranges spanning several orders of magnitude, from around 0.03 to 10 micrograms of loaded protein, which is far wider than what traditional staining achieves.5PubMed. Real-time and quantitative protein detection via polyacrylamide gel electrophoresis and online intrinsic fluorescence imaging Still, most labs use conventional staining, so checking the linear range of your specific setup matters.
When One Plasmid Gives You Three Bands
New gel readers are often confused when a single plasmid DNA prep produces two or three distinct bands instead of one. This is normal. Plasmid DNA exists in at least three structural forms: supercoiled (compact and tightly wound), nicked circular (one strand cut, so the coil relaxes into a floppy open circle), and linear (both strands cut at the same spot). Each shape moves through the gel differently because their effective sizes differ even though they contain the same number of base pairs.
Supercoiled DNA is compact and typically migrates fastest, appearing as the lowest band. Linear DNA migrates at an intermediate rate. Nicked circular DNA is the bulkiest shape and usually runs the slowest, sitting closest to the wells. However, this order is not set in stone. The relative positions of supercoiled and linear forms can reverse depending on the agarose concentration and the molecular weight of the plasmid. At lower gel concentrations, supercoiled DNA may actually migrate more slowly than the linear form, and at higher concentrations the relationship flips.6Nucleic Acids Research. Physical properties and gel electrophoresis behavior of R12-derived plasmid DNAs The nicked circular form is always the most retarded of the three.7PubMed. Resolution of circular, nicked circular and linear DNA, 4.4 kb in length, by electrophoresis in polyacrylamide solutions
This matters in practice because a good-quality plasmid prep should show mostly supercoiled DNA. If the nicked or linear bands are disproportionately bright, the sample has been damaged, whether by mechanical shearing, nuclease contamination, or rough handling during purification. The relative brightness of the three bands is a quick quality check before you use the plasmid for cloning, transfection, or other downstream work.
The staining dye you use can also shuffle band order. Certain intercalating dyes unwind the DNA helix, which changes the effective compactness of supercoiled molecules and therefore changes how they migrate. One study found that the elution order of linear, supercoiled, and nicked forms changed depending on which dye was present during the run.8PubMed. Electrophoretic behavior of plasmid DNA in the presence of various intercalating dyes The broader point here applies to all gel reading: the dye you choose is not cosmetically neutral. Different commercial DNA-binding dyes vary in sensitivity and in how much they alter the electrophoretic mobility of your nucleic acids.9PubMed. Comprehensive assessment of 12 commercial DNA-binding dyes as alternatives to ethidium bromide for agarose gel electrophoresis
Reading RNA Gels
When you run total RNA on a gel, the most prominent features are two bright bands corresponding to the 28S and 18S ribosomal RNA subunits. These are the workhorses of the ribosome and by far the most abundant RNA species in a cell, so they dominate the gel image. Below them you might see a faint band for 5.8S or 5S rRNA, and the messenger RNA appears as a faint smear in the background rather than as discrete bands, because mRNAs come in many different sizes.
The classic rule of thumb for RNA quality is that the 28S band should be about twice as bright as the 18S band (a 2:1 ratio), indicating intact RNA. If the two bands are equally bright, or the 18S band is brighter, or both bands are faint against a heavy background smear, the RNA is degraded. Degradation chews up the longer 28S molecules first, shifting the ratio downward and releasing fragments that spread out as a diffuse smear across the lane. An elevated baseline between and below the ribosomal bands is another hallmark of breakdown.
That said, the 28S-to-18S ratio is a rougher guide than it seems. When researchers measured the ratio repeatedly from the same samples in independent runs, the variability was surprisingly high, with coefficients of variation ranging from 19 to 24 percent. Part of the problem is that calculating the ratio depends on where you define the start and end of each peak, which is somewhat subjective, especially in automated systems. Because of these limitations, more modern approaches use algorithms that assess the entire electropherogram trace rather than just the two ribosomal peaks, producing a single integrity score. But if you are reading a traditional gel by eye, looking for two sharp ribosomal bands with the upper one clearly brighter and minimal smearing below them remains a solid first-pass quality check.
Interpreting Protein Gels
For proteins, the standard approach is SDS-PAGE. The detergent SDS coats each protein with a uniform negative charge proportional to its length, which means proteins separate primarily by size just like DNA does in agarose gels. You load a protein molecular weight marker in one lane and compare your sample bands to it, reading off approximate sizes in kilodaltons.
A clean protein gel with well-separated, sharp bands is the ideal. In reality, bands can appear fuzzy, doubled, or at unexpected positions. A band running higher than the predicted molecular weight might indicate that the protein is glycosylated, since sugar chains add mass. A band that appears as a doublet could mean partial proteolytic cleavage, incomplete reduction of disulfide bonds, or post-translational modifications like phosphorylation. Phosphorylated proteins, for example, can migrate differently from their unphosphorylated counterparts, sometimes producing a distinct upward-shifted band alongside the main one.
One thing that catches people off guard is the difference between denaturing and native gels. In standard SDS-PAGE, proteins are boiled in detergent and reducing agent, which unfolds them into linear chains. Migration then reflects the polypeptide’s molecular weight alone. In native PAGE, no SDS is added and the protein keeps its natural folded shape and any subunit associations. A protein that runs as a single band at, say, 35 kDa under denaturing conditions might appear at 140 kDa under native conditions because it normally forms a four-subunit complex in the cell.10PubMed Central. Native SDS-PAGE: High Resolution Electrophoretic Separation of Proteins With Retention of Native Properties Including Bound Metal Ions If you see a band at a much higher molecular weight than you expected, check whether the gel was run under native conditions, because that band may represent the intact protein complex rather than the individual subunit.
Two-dimensional approaches combine both methods by running a native gel first, then cutting out a lane and laying it across the top of a denaturing gel run perpendicular to the first. This reveals which subunits belong to which complexes, a strategy used in studies of multi-protein signaling machinery.11PubMed. Two dimensional Blue Native-/SDS-PAGE analysis of SLP family adaptor protein complexes
Shifted Bands in Mobility Shift Assays
A specialized type of gel reading involves the electrophoretic mobility shift assay, or EMSA. You mix a labeled piece of DNA or RNA with a protein you suspect binds to it, then run the mixture on a native gel. If the protein binds the nucleic acid, the resulting complex is bulkier than the free nucleic acid alone and migrates more slowly. On the gel, this shows up as a new band above the position of the free probe.12Nature Protocols. Electrophoretic mobility shift assay (EMSA) for detecting protein–nucleic acid interactions The free probe band either diminishes in intensity or disappears entirely if the protein saturates all the available nucleic acid.
Reading an EMSA gel is about watching for that upward shift. If you see it, the interaction is confirmed. If you add increasing amounts of protein and the shifted band gets brighter while the free probe band gets dimmer, you are looking at dose-dependent binding. Adding an unlabeled competitor nucleic acid that competes away the protein should cause the shifted band to fade and the free probe band to return, verifying that the interaction is specific.13PubMed. Unraveling biomolecular interactions: a comprehensive review of the electromobility shift assay Sometimes you also see a “super-shift” if you add an antibody against the protein: the antibody-protein-DNA complex is even bigger and migrates even more slowly, producing a third band above the first shifted band.
Common Artifacts and What They Mean
Not every mark on a gel is meaningful data. Learning to spot artifacts saves hours of chasing phantom results.
- Smearing: A lane with a diffuse smear instead of crisp bands usually indicates degradation. For DNA, this means nuclease contamination or mechanical shearing. For RNA, it signals enzymatic breakdown. A smear that runs from the well downward without any distinct bands suggests the sample was severely degraded before loading.
- Smile effect: Bands that curve upward at the edges of the gel, forming a smile shape, come from uneven heating during the run. The center of the gel gets hotter than the edges, causing molecules in the center lanes to migrate faster. This thermal distortion is a well-known problem in electrophoresis and can be minimized by running gels at lower voltage or using a buffer-recirculation system to distribute heat evenly.14ELECTROPHORESIS. Thermal instability of electrophoresis gels
- Well artifacts: Bright material stuck in or near the loading well usually means the sample contained large aggregates, genomic DNA contamination (in a plasmid prep), or protein that precipitated before entering the gel.
- Ghost bands: Faint bands that appear in lanes where you did not load that particular sample can result from spillover during loading or from the staining dye bleeding between lanes. Careful pipetting and proper well spacing help avoid this.
- Uneven band intensity across lanes: If you loaded equal amounts in every lane but some bands appear brighter, the most common culprits are uneven gel thickness (caused by an improperly leveled casting tray) or inconsistent pipetting.
Pulsed-Field Gels for Very Large DNA
Conventional agarose gels lose their resolving power for DNA fragments above about 50 kilobases. Very large molecules get stuck in the pores or move at the same rate regardless of size, making it impossible to distinguish, say, a 200 kb fragment from a 500 kb one. Pulsed-field gel electrophoresis (PFGE) solves this by alternating the direction of the electric field during the run. Each time the field switches, the DNA molecules have to reorient themselves before they can start migrating again, and larger molecules take longer to reorient. This size-dependent reorientation time separates fragments that would otherwise co-migrate, allowing resolution of DNA in the megabase range.15PubMed Central. Pulsed-Field Gel Electrophoresis
Reading a PFGE gel follows the same general logic as reading any other gel: you compare sample bands to a ladder of known-size fragments. The ladder for PFGE, though, is made from chromosomal DNA of organisms whose chromosome sizes are well characterized, such as yeast chromosomes that range from a few hundred kilobases to over a megabase. PFGE is commonly used for bacterial strain typing in epidemiology, because digesting a bacterium’s chromosome with a rare-cutting restriction enzyme produces a distinctive pattern of large fragments. Two isolates with identical banding patterns are almost certainly the same strain, while differences in even one or two bands suggest they are related but not identical.
Choosing and Trusting Your Staining Dye
Ethidium bromide has been the default DNA stain for decades, but safety concerns have driven labs toward alternatives like SYBR Safe, SYBR Green, GelRed, and others. These dyes differ in more than just toxicity profiles. A comprehensive comparison of twelve commercial DNA-binding dyes found that they varied in both sensitivity and in their effect on how DNA migrates through the gel.9PubMed. Comprehensive assessment of 12 commercial DNA-binding dyes as alternatives to ethidium bromide for agarose gel electrophoresis Some dyes shift bands slightly, making them appear to be a different size than they actually are when compared to a ladder stained differently or run without dye.
The practical takeaway is to stain your ladder and your samples the same way. If you are adding dye to the gel before casting (pre-cast staining), both the ladder and the samples experience the same mobility shift, so your size estimates stay accurate. Problems arise when you stain after the run (post-staining) and compare to a manufacturer’s size chart that was generated under different conditions, or when you mix pre-cast and post-stained lanes on the same gel. Consistency in staining protocol matters as much as consistency in gel concentration and running buffer.