template versus coding strand

DNA has two strands wound together, but when a gene is transcribed, RNA polymerase reads only one of them. So which one? And why does the RNA end up looking like the strand that was never read? This entry untangles the two roles the two strands play.

Of the two DNA strands, the one the polymerase actually reads is the template strand (also called the antisense strand). The polymerase builds RNA complementary to it, base-pairing A-to-U and G-to-C. But here is the satisfying twist: because base-pairing is symmetric, the RNA turns out identical in sequence to the OTHER strand — the coding strand (or sense strand) — with the single difference that RNA carries U wherever DNA had T. So if the coding strand reads 5'-ATGCCT-3', the RNA reads 5'-AUGCCU-3', and the template strand (the one read) is its complement, 3'-TACGGA-5'. We write gene sequences as the coding strand precisely because it 'reads like' the RNA. Which strand serves as template is not fixed for the whole chromosome — it is decided gene by gene, since neighbouring genes can be transcribed in opposite directions, each using whichever strand points the right way.

Getting these names straight clears up a classic source of confusion. The template strand is the one with the information being read, yet the coding strand is the one whose sequence we quote when we talk about codons and the genetic code. When you later see a gene's 'sequence' in a database and find the start codon ATG sitting right there, you are looking at the coding strand — the polymerase never touched it; it read its mirror image.