17 DECEMBER 2021 • VOL 374 ISSUE 6574 1427Early results were humbling, with median
scores below 60. But over time, the model-
ers learned tricks to improve their calcula-
tions. For example, stretches of amino acids
shared by two proteins often fold similarly. If
a protein with an unknown structure shares,
say, 50% of its amino acid sequence with a
protein that does have a known structure, the
latter can serve as a “template” to guide the
computer models.
Another major insight came from evolu-
tion. Investigators realized that if one amino
acid changed in a protein shared by closely
related organisms, like chimpanzees and
humans, amino acids located nearby in the
folded molecule would have to change, too,
to preserve the protein’s shape and function.
That means investigators can narrow down
a protein’s shape by looking for amino acids
that coevolve: Even if they are far apart on
the unfolded chain, they are likely neighbors
in the final 3D structure.
By 2018, the modelers were often scor-
ing in the mid-70s. Then, AlphaFold, an AI-
driven software program, entered the scene.
The program, developed by Google sister
company DeepMind, trains itself on data-
bases of experimentally solved structures.
In its first competition, its median score
was close to 80, and it won 43 of 90 matches
against other algorithms. In 2020, its suc-
cessor, AlphaFold2, shone even brighter.
Powered by a network of 182 processors op-
timized for machine learning, AlphaFoldrang up a median score of 92.4—on par with
experimental techniques.
“I never thought I’d see this in my lifetime,”
John Moult, a structural biologist at the Uni-
versity of Maryland, Shady Grove, and CASP
co-founder, said at the time.
This year, AI predictions shifted into over-
drive. In mid-July, Baker and his colleagues
reported that their AI program RoseTTA-
Fold had solved the structures of hundreds
of proteins, all from a class of common
drug targets. A week later, DeepMind scien-
tists reported they had done the same for
350,000 proteins found in the human body—
44% of all known human proteins. In coming
months, they expect their database will grow
to 100 million proteins across all species,
nearly half the total number believed to exist.
The next step is to predict which of those
proteins work together and how they inter-
act. DeepMind is already doing just that. In
an October preprint, its scientists unveiled
4433 protein-protein complexes, revealing
which proteins bind to one another—and
how. In November, RoseTTAFold added an-
other 912 complexes to the tally.
Code for AlphaFold2 and RoseTTAFold is
now publicly available, helping other scien-
tists jump into the game. In November, re-
searchers in Germany and the United States
used AlphaFold2 and cryo-EM to map the
structure of the nuclear pore complex, an as-
sembly of 30 different proteins that controls
access to the cell nucleus. In August, Chineseresearchers used AlphaFold2 to map the
structures for nearly 200 proteins that bind
to DNA, which could be involved in every-
thing from DNA repair to gene expression.
Last month, Google’s parent company, Al-
phabet, launched a new venture that will use
predicted protein structures to design new
drug candidates. And Baker’s team is using
its software to dream up novel protein se-
quences that will fold into stable structures,
an advance that could lead to new antivirals
and catalysts.
Even now, scientists studying SARS-
CoV-2 are using AlphaFold2 to model the
effect of mutations in the Omicron variant’s
spike protein. By inserting larger amino
acids into the protein, the mutations have
changed its shape—perhaps enough to keep
antibodies from binding to it and neutral-
izing the virus.
Much work remains. Protein structures
aren’t static; they bend and twist as they do
their jobs, and modeling those changes re-
mains a challenge. And it’s still a daunting
task to visualize most of the large, multi-
protein complexes that carry out myriad jobs
in cells. But this year’s explosion of AI-driven
advances offers a view of the dance of life as
never seen before, a panorama that will for-
ever change biology and medicine. jAI-powered
predictions reveal
the shapes of proteins
by the thousands
VIDEO AND PODCAST
The year in Science and our People’s
Choice award: science.org/botySCIENCE science.org