Gel Electrophoresis: Separating Molecules by Size

A uniformly charged molecule, an uneven obstacle course

DNA's phosphate backbone carries a negative charge that is essentially uniform per unit length — a 500 base-pair fragment has almost exactly twice the charge and twice the mass of a 250 base-pair fragment, which means the charge-to-mass ratio barely changes with size. In free solution under an electric field this would make all DNA fragments migrate at nearly the same speed, which would be useless for sizing anything. Gel electrophoresis solves this by forcing the DNA to move through a sieving matrix — a porous gel, typically agarose or polyacrylamide — where size, not charge, becomes the thing that actually limits speed.

live demo · charged fragments migrating through a gel mesh, sorted by size● LIVE

Two sieving regimes: Ogston and reptation

For small-to-medium fragments moving through a gel whose pore size is comparable to or larger than the molecule, the Ogston sieving model applies: the gel behaves like a random sieve, and a larger, more globular molecule simply has fewer pore pathways available to it, so mobility falls off roughly with the molecule's size relative to the average pore size. For larger DNA fragments (roughly above 1-2 kb in a typical agarose gel), the DNA is bigger than the gel's pores and instead has to snake head-first through the mesh in an elongated, worm-like configuration — a regime called reptation (from the Latin for "to creep"), borrowed from polymer physics. In the reptation regime the relationship between mobility and molecular size flattens out, which is why very large fragments become progressively harder to resolve on a standard gel and why techniques like pulsed-field gel electrophoresis exist to separate chromosome-sized DNA by periodically switching the field direction.

The calibration curve: turning distance into size

Within the regime where a gel actually resolves fragments well, mobility relates to fragment size in a form that is close to log-linear over a useful range:

distance migrated ~ log10(fragment length in base pairs)     // approximately, over the gel's resolving range

This is exactly why every gel is run with a DNA ladder — a lane loaded with fragments of known, standard sizes — alongside the unknown samples. Plotting the ladder's known sizes (on a log scale) against migration distance gives a calibration curve; reading an unknown sample's migration distance off that same curve, run under identical conditions in the identical gel, gives its estimated size. The calibration only holds within the range spanned by the ladder and only for fragments that are roughly the same shape (linear, double-stranded DNA behaves differently from circular plasmid DNA of the same length, for instance, because shape affects how the molecule threads through the mesh).

Why the gel concentration matters

A higher percentage agarose gel has smaller average pore size, which resolves small fragments better but makes large fragments crawl so slowly they barely separate from each other; a lower percentage gel has larger pores, resolving large fragments well but letting small fragments all run together near the dye front. Choosing gel percentage is choosing which size range of fragments you actually want the sieving effect to discriminate between — a typical general-purpose agarose gel runs around 0.7-2%, with the low end for large fragments (tens of kb) and the high end for small ones (under a few hundred bp).

SDS-PAGE: forcing proteins to obey the same rule

Proteins do not have DNA's convenient uniform charge-per-length, since their native charge depends on their specific amino acid sequence, and native folding gives different proteins wildly different shapes at the same mass. SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) fixes both problems at once: SDS is an anionic detergent that coats a denatured, unfolded protein roughly in proportion to its length, swamping the protein's native charge with a strongly negative, size-proportional charge, and denaturation removes the shape variability. The result is that SDS-PAGE, like DNA gel electrophoresis, ends up sorting almost purely by size (molecular weight) rather than by native charge or shape — the same sieving-plus-uniform-charge trick, applied to a different molecule.

Frequently asked questions

Why doesn't DNA just separate by charge in an electric field without a gel?

DNA's charge is nearly proportional to its length, so the charge-to-mass ratio barely changes with fragment size — in free solution, most fragments would migrate at almost the same speed. The gel's porous mesh is what makes size the limiting factor, since smaller fragments thread through the pores faster than larger ones.

What is a DNA ladder for?

It is a reference lane of fragments with known sizes run alongside the samples on the same gel. Because migration distance is roughly log-linear with fragment size within the gel's resolving range, plotting the ladder's known sizes against their migration distances gives a calibration curve for reading the size of unknown fragments in other lanes.

Why does SDS-PAGE separate proteins by size when native charge varies so much between them?

SDS is a detergent that binds denatured protein roughly in proportion to its length, which swamps each protein's native charge with a new, size-proportional negative charge, while denaturation also removes shape differences. That combination makes SDS-PAGE mobility depend almost entirely on molecular weight, the same way gel sieving makes DNA mobility depend mostly on fragment length.

Try it live

Everything above runs in your browser — open Gel Electrophoresis and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Gel Electrophoresis simulation

Gel Electrophoresis: How a Gel Turns Charge Into a Size Sort