if the four basic nucleobases decide gene outcome (ACGT), but have strict complementary base pairing (A-C & G-T), are there then four values to variate with or just two, like binary?

Question

**Correction**: A-T & C-G, not ~~A-C & G-T~~
The question arose from [here](https://www.reddit.com/r/Showerthoughts/comments/k3b4ba/just_like_computers_speak_in_binary_1s_and_0s_the/ge2jzil?utm_source=share&utm_medium=web2x&context=3), where I compared binary code to human DNA’s ACGT, then someone who knows more about IT than me made a good question.

If it’s not clear what I’m asking yet:
in binary you can have a lot of variation with the values 1 and 0
in human DNA (if we ignore RNA and (U)racil for now) there are four values, A, C, G, T. But since they bind strictly A-C & G-T doesn’t that technically leave DNA with only two factors to variate with?

Like: ~~A-C, A-C, G-T, A-C~~ A-T, A-T, C-G, A-T
Instead of: AGTATAC

In: Biology

Answer 1

When replication happens, only one side of the helix gets copied.

The two ends of the helix are unique from one another, the body can tell them apart, and has a preferred direction that it will follow.

Because of this, even though they always come in pairs, swapping a pair would change the message being copied, because of this preferred direction.