Coevolution Contact Map: How AI Reads Structure from Sequence (2D)

Sequence Alignment

MSA depth (sequences) 150

Mutation noise 35%

Top-K contacts used 24

Show hidden true contact map

Live Readouts

Top-K precision–

Top-K recall–

Mean MI of top-K (bits)–

True contacts (hidden)–

Computed MI heat (both panels) Top-K pick, true positive Top-K pick, false positive Hidden true contact (right panel)

How it works

Real proteins that fold correctly show coevolution: if residue i mutates, a spatially contacting residue j often mutates too, to keep the pair chemically compatible. Across a deep multiple sequence alignment (MSA) of homologous sequences, this leaves a statistical footprint even though i and j can be far apart along the chain.

Mutual information between columns i, j:
MI(i,j) = Σ_a Σ_b p(a,b) · log₂[ p(a,b) / (p(a)·p(b)) ]

a, b = residue states observed at columns i, j
p(a,b) = joint frequency across the M sequences
High MI ⇒ i and j are statistically coupled ⇒ likely in 3D contact

This simulator builds a hidden helix-turn-helix protein backbone (32 residues, never drawn) purely to define ground-truth 3D contacts, then synthesizes an MSA where a subset of those true contact pairs carry a coevolution signal, corrupted by the noise slider. From only the generated sequences — never the hidden structure — it computes real pairwise mutual information for every one of the 406 eligible residue pairs (|i−j| ≥ 4) and renders the full inferred contact-map heatmap live on the left. The right panel reveals the hidden true contact map for comparison, and the top-K highest-MI pairs are outlined amber (true positive) or red (false positive) on both.

MSA depth — more sequences average out sampling noise, so MI estimates converge to the true coupling and precision/recall rise (just as real structure prediction needs deep alignments).
Mutation noise — degrades both the background signal and the coevolutionary coupling strength; push it high enough and even a deep MSA can't recover the map.
Top-K — how many of the highest-MI pairs are trusted as contacts; too few misses real contacts (low recall), too many admits false positives (low precision).