Sanger (dideoxy chain-termination) sequencing copies a single-stranded template with DNA polymerase in a mix that contains normal dNTPs plus a small fraction of fluorescently-labeled dideoxynucleotides (ddNTPs), split across four tubes/lanes — one per base. A ddNTP lacks the 3'-OH needed to add the next base, so any strand that incorporates one stops growing right there. Within each lane's population of template copies this produces a truncated fragment ending at every position where that base occurs in the synthesized strand:
lane(base b) = { fragment i : synthesized_base(i) = b }
fragment i = primer + synthesized bases 1..i, length = i, color = dye(base_i)
The four reactions are loaded side-by-side into a gel (or, in modern instruments, one capillary) and pulled through a sieving matrix by an electric field. DNA is uniformly negatively charged, so separation happens only because the matrix drags longer fragments more (the reptation model):
electrophoretic mobility μ(N) ∝ 1/N
migration velocity v(N) = k · E / N
time to reach detector t(N) = L / v(N) → grows roughly ∝ N
Small fragments outrun large ones down every lane, so a detector line low in the gel is crossed shortest-fragment-first. Reading off which lane (i.e. which base) each successive crossing belongs to, in arrival order, reconstructs the synthesized strand base-by-base — and taking the complementary base at each step reconstructs the original template. Thermal diffusion adds a small random jitter to every fragment's position (band broadening), which is why very long, closely-sized fragments start to overlap and become harder to call — the peak spacing Δt readout tracks this shrinking gap live.
- Template length — how many bases (and therefore fragments) are in this run.
- Electric field strength — raises migration speed but also raises diffusive broadening (Joule heating), the real speed/resolution trade-off operators tune on a real sequencer.
- New Template + Run — draws a fresh random template and restarts electrophoresis.
- Custom template — type your own A/C/G/T sequence and watch the simulator derive and separate its real termination fragments.