Cambridge, early 1953. James Watson and Francis Crick are racing to find the shape of DNA, the molecule that carries genes. X-ray pictures say it is a helix, the same width all the way along. Inside it sit four kinds of base, A, G, T and C, in any order.
A jigsaw, then, with four kinds of piece and no picture on the box. How do the pieces fit across the middle without the width ever changing? And how could such a thing copy itself?
The cardboard
Watson cut the four bases out of stiff card, true to scale, and pushed them around his desk. Like with like bulged. Then, on the morning of Saturday 28 February, he saw it: A with T has exactly the shape of G with C. Two pairs, one width. Five weeks later he and Crick sent their letter to Nature.
Now it's your desk
This game lets you be Watson, with nothing but some cardboard. The pieces are his cut-outs. One chain is given; find the other, and watch whether the rails stay straight. Then pull your ladder apart and see what he and Crick saw next: it can copy itself.
Story: Watson, The Double Helix (1968), ch. 26. The letter itself does not mention cardboard.
Watson's Cardboardthe desk
Build the other chain so every rung holds when you zip it.
Two blue rails, his cut-out pieces between them. Some fit, some don't. No clock. Two ladders: the first to find out, the second to score.
(same number, same chains)
Build the other chain so every rung holds when you zip it.
Your ladder
Piecestap one to pin it
Build the other chain so every rung holds when you zip it.
0/100
How to play
Build the other chain so every rung holds when you zip it.
The ladder. Two blue rails (the backbones of DNA) with rungs between them. Each rung is two pieces, one pinned to each rail, meeting in the middle.
The pieces. Four kinds: A and G are long, T and C are short. Each has an end with two contact points, top and bottom; each point is a bump or a dent.
A rung holds when the two pieces exactly span the gap between the rails and every bump meets a dent. Too long and the rails bow out. Too short and they pinch in. A bump against a bump pushes the piece away.
The ladder
One rail already has its pieces. Tap a piece in the tray to pin it in the next empty spot (marked in red); tap a pinned piece to take it back. Nothing is tested until you press Zip it. Then every rung is tested, top to bottom. There is no clock: take as long as you like.
There are two ladders. The first is for finding out: the red rungs tell you what went wrong. The second is a new chain, and it is the one that scores.
Your score
Ten points for every rung of your second ladder that holds, out of 100, with your first ladder's score beside it. Next to it: what someone pinning at random gets (about 25), someone who reads only the lengths or only the ends (about 50), and someone who reads both (100). Then what your ladder shows, your letters counted, and your ladder copied and twisted into the real shape. Play again for a new chain.
What's real here
The game in one line
You build the second chain of a DNA ladder from cut-out pieces, zip it to see which rungs hold, and get a score out of 100: ten points a rung. The end screen shows your ladder copied and twisted into the helix. The rule that decides which rungs hold is Watson and Crick's; the pieces, their lengths and the ladder are this game's.
The cardboard
The pieces are cardboard because that is how the pairs were found: in February 1953 Watson cut the four bases out of stiff card, to scale, and pushed them around his desk until an adenine–thymine pair and a guanine–cytosine pair turned out to be the same shape (Watson, The Double Helix, 1968, ch. 26). The cardboard is history, not the letter, which says only that the structure rests on "published experimental data and stereochemical arguments". The aim, to build the other chain, is the letter's own consequence: "if the sequence of bases on one chain is given, then the sequence on the other chain is automatically determined."
What the paper found
In a one-page letter to Nature in April 1953, James Watson and Francis Crick proposed the structure of DNA. Two chains wind round the same axis, right-handed, running in opposite directions. The sugar-phosphate backbones are on the outside (Linus Pauling's model of a month earlier had three chains with the phosphates in the middle). The bases are inside, flat, and joined across in pairs by hydrogen bonds. One base of each pair must be a purine (adenine or guanine, two rings) and the other a pyrimidine (thymine or cytosine, one ring). Assuming the bases take their most plausible (keto) forms, only adenine with thymine and guanine with cytosine can bond. So the order of bases along one chain is unrestricted, but it fixes the order on the other. That explains why, as Erwin Chargaff had measured, DNA holds as much adenine as thymine and as much guanine as cytosine. And, in the letter's most famous sentence, "It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material." The helix has a base pair every 3.4 Å along the axis, each turned 36° from the last, so it repeats every 10 pairs (34 Å); the phosphates sit 10 Å from the axis. The authors say the model is "roughly compatible" with the X-ray data then published and "must be regarded as unproved" until checked.
Mechanism: a toy
This game demonstrates that the letter's two conditions, one purine with one pyrimidine (so every rung has the same width) and hydrogen bonds that line up (in the keto forms), leave exactly two pairs, and that this forces one chain to spell out the other, so a ladder pulled apart can be rebuilt twice over. Each condition alone would leave twice as many pairs; together they leave A–T and G–C.
It assumes pieces instead of molecules. Length stands for ring size: a long piece (A, G) and a short piece (T, C) exactly span the rails; two long pieces are too wide, two short ones too narrow. The lengths are our units; any lengths with purines longer than pyrimidines give the same answer. The bumps and dents stand for hydrogen donors and acceptors at the letter's positions 6 and 1. Which base has which is standard chemistry for the keto forms, not printed in the letter. The game uses the letter's two contacts per pair; guanine and cytosine were later found to share a third. The ladder and its sequence (drawn from the game number) are ours. The end screen compares your ladder with other ways of playing, from our simulation: pinning at random holds about 2 or 3 rungs of 10, reading only the lengths or only the ends about 5, reading both all 10; the pattern of failures on your ladder (only too wide or too narrow, only bump on bump, or both) says which clue was missing.
The paper establishes a proposed structure and the pairing argument behind it, from published data and stereochemistry; it says the structure is unproved. It does not describe how copying happens: the letter only says the pairing "suggests" a mechanism. The copy in this game (each half rebuilding its partner) is the version Watson and Crick set out a month later, in Nature 171:964, and that later experiments bore out. Neither the zip nor the copy in this game is evidence for the structure: the game shows what follows if it is right.
What is cut
Pauling's and Fraser's rival models; the argument that the material is a salt and that the phosphates outside give cations easy access; the water content; the reason ribose would not fit; the helix's geometry, except at the end, where the copies twist a tenth of a turn per rung, the letter's 36°, one full turn per ten rungs.
The paper
J. D. Watson and F. H. C. Crick, "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid", Nature 171:737–738 (1953). doi:10.1038/171737a0. Read it at nature.com/articles/171737a0.
Game built from this paper; the paper is the authority where the two disagree. Fonts: Permanent Marker (Apache License 2.0) and Courier Prime (SIL Open Font License).