Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0203
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-07-13T21:53:15.138572Z
status: published
lens: g_space
sector: science
headline: Highly accurate protein structure prediction with AlphaFold
dek: Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous experimental effort 1-4 , the structures of around 100,000 unique proteins have been determined 5 , but this represents a small fraction of the billions of known protein sequences 6,7 . Structural coverage is bottlenecked by the months to years of painstaking effort required to determine a single protein structure. Accurate computational approaches are needed to addr…
gain_title: AlphaFold provides a computational method that regularly predicts protein three-dimensional structures with atomic accuracy from amino acid sequence alone, even without homologous structures.
problem_title: (none)
trace_subject: (none)
gain_reading: AlphaFold provides a computational method that regularly predicts protein three-dimensional structures with atomic accuracy from amino acid sequence alone, even without homologous structures.
gain_evidence: can regularly predict protein structures with atomic accuracy even in cases in which no similar structure is known | Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence
problem_reading: (none)
problem_evidence: (none)
quick_read: On July 15, 2021, Nature published the AlphaFold study describing a redesigned neural network that predicts the three-dimensional structure a protein will adopt based solely on its amino acid sequence. The authors reported validation in CASP14, where the model regularly achieved atomic accuracy even when no homologous structure was available and performed competitively with experimental structures.

The result matters because experimental determination of protein structures is slow and covers only a small fraction of known sequences, limiting mechanistic understanding of function and large-scale structural bioinformatics. A computational method with atomic accuracy could expand structural coverage dramatically, though the source does not yet detail performance across all protein classes, experimental validation beyond CASP14, or downstream biological applications.
limitation: 
tag: Evidence-backed gain
key_points: AlphaFold was validated in CASP14, the 14th Critical Assessment of protein Structure Prediction, showing accuracy competitive with experimental structures. | The approach incorporates physical and biological knowledge and leverages multi-sequence alignments in a redesigned neural network. | Experimental structure determination covers around 100,000 unique proteins versus billions of known sequences, creating a coverage bottleneck.
rundown: The paper describes an entirely redesigned neural network-based model that incorporates physical and biological knowledge about protein structure and multi-sequence alignments into the deep learning algorithm.

Structural coverage is limited because determining a single protein structure requires months to years of effort, while only around 100,000 unique proteins have been experimentally determined out of billions of known sequences.

In CASP14 validation, the method greatly outperformed other methods and achieved accuracy competitive with experimental structures in a majority of cases, including cases with no similar structure known.
sources:
- peer_reviewed | Nature | https://doi.org/10.1038/s41586-021-03819-2 | 2021-07-15
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

611e6a919f115a84c7c86687044cb2358bfd82411f1d1c98333c9184e9339d37

previous

0000000000000000000000000000000000000000000000000000000000000000