A survey of published research, from the Essentia Memoria newsroom. It lists real papers, each looked up at its DOI registry; the game has not repeated their results.
Research survey
What AI research has found: proteins and the molecules of life
Happened: · Published:
12 published works on proteins and the molecules of life, each looked up at its DOI registry before it was listed here, with what each one reports in plain words.
Watch
Explanatory visualisation, drawn from the record's numbers. It is not footage of the event: nobody filmed it.
Listen
Transcript
News from the Essentia Memoria game universe. What AI research has found: proteins and the molecules of life. 12 published works on proteins and the molecules of life, each looked up at its DOI registry before it was listed here, with what each one reports in plain words. This survey lists what each paper reports. The game has not repeated these results. A search finds papers that match its words, so a listed paper may use AI only in part. The search is limited to the fifty most-cited matches, so the newest results may be missing until they are cited. This report was written from the world record by a fixed template. The full story and its source records are on essentiamemoria.com.
The question
How did AI learn to predict the shape a protein folds into?
Claims, and where each one comes from
- Where this research lives in the game (a place the world grammar builds)laboratory → laboratory
Game simulation · Set in code
src/lib/world-grammar/place-grammar.ts · resolvePurpose('laboratory') - Published works in this survey, each resolved at its DOI registry12
Real-world knowledge · Source [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
- AlphaFold predicts the three-dimensional shape of a protein from its amino-acid sequence, often close to laboratory accuracy.Jumper 2021
Real-world knowledge · Source [1]
- Harnessing protein folding neural networks for peptide–protein dockingTsaban 2022
Real-world knowledge · Source [2]
- [31] PHD: Predicting one-dimensional protein structure by profile-based neural networksRost 1996
Real-world knowledge · Source [3]
- Evolutionary-scale prediction of atomic-level protein structure with a language modelLin 2023
Real-world knowledge · Source [4]
- ProtGPT2 is a deep unsupervised language model for protein designFerruz 2022
Real-world knowledge · Source [5]
- P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structureKrivák 2018
Real-world knowledge · Source [6]
- Accurate prediction of protein structures and interactions using a three-track neural networkBaek 2021
Real-world knowledge · Source [7]
- Improved protein structure prediction using potentials from deep learningSenior 2020
Real-world knowledge · Source [8]
- Highly accurate protein structure prediction for the human proteomeTunyasuvunakool 2021
Real-world knowledge · Source [9]
- AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequencesVáradi 2023
Real-world knowledge · Source [10]
- Improved protein structure prediction using predicted interresidue orientationsYang 2020
Real-world knowledge · Source [11]
- Prediction of protein folding class using global description of amino acid sequence.Dubchak 1995
Real-world knowledge · Source [12]
References, checked at their registry
[1] Jumper (2021). Highly accurate protein structure prediction with AlphaFold. Nature. doi:10.1038/s41586-021-03819-2
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Highly accurate protein structure prediction with AlphaFold” (2021)
[2] Tsaban (2022). Harnessing protein folding neural networks for peptide–protein docking. Nature Communications. doi:10.1038/s41467-021-27838-9
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Harnessing protein folding neural networks for peptide–protein docking” (2022)
[3] Rost (1996). [31] PHD: Predicting one-dimensional protein structure by profile-based neural networks. Methods in enzymology on CD-ROM/Methods in enzymology. doi:10.1016/s0076-6879(96)66033-9
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “[31] PHD: Predicting one-dimensional protein structure by profile-based neural networks” (1996)
[4] Lin (2023). Evolutionary-scale prediction of atomic-level protein structure with a language model. Science. doi:10.1126/science.ade2574
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Evolutionary-scale prediction of atomic-level protein structure with a language model” (2023)
[5] Ferruz (2022). ProtGPT2 is a deep unsupervised language model for protein design. Nature Communications. doi:10.1038/s41467-022-32007-7
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “ProtGPT2 is a deep unsupervised language model for protein design” (2022)
[6] Krivák (2018). P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure. Journal of Cheminformatics. doi:10.1186/s13321-018-0285-8
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure” (2018)
[7] Baek (2021). Accurate prediction of protein structures and interactions using a three-track neural network. Science. doi:10.1126/science.abj8754
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Accurate prediction of protein structures and interactions using a three-track neural network” (2021)
[8] Senior (2020). Improved protein structure prediction using potentials from deep learning. Nature. doi:10.1038/s41586-019-1923-7
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Improved protein structure prediction using potentials from deep learning” (2020)
[9] Tunyasuvunakool (2021). Highly accurate protein structure prediction for the human proteome. Nature. doi:10.1038/s41586-021-03828-1
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Highly accurate protein structure prediction for the human proteome” (2021)
[10] Váradi (2023). AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Research. doi:10.1093/nar/gkad1011
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences” (2024)
[11] Yang (2020). Improved protein structure prediction using predicted interresidue orientations. Proceedings of the National Academy of Sciences. doi:10.1073/pnas.1914677117
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Improved protein structure prediction using predicted interresidue orientations” (2020)
[12] Dubchak (1995). Prediction of protein folding class using global description of amino acid sequence.. Proceedings of the National Academy of Sciences. doi:10.1073/pnas.92.19.8700
Verified: the work exists and matches · checked at Crossref on Oct 28, 2026, 10:25 PM UTC · the registry holds “Prediction of protein folding class using global description of amino acid sequence.” (1995)
Review
Not peer reviewed. No human scientist has reviewed this survey. Its template and landmark summaries were written with an AI assistant (Claude); the papers it lists are real and each was checked at its registry, so anyone can follow the links and read them.
- Each listed DOI was resolved, and its title, year and first author matched the listing.
- Each work the search found is held by Crossref, a registry independent of the search.
- Each number in the headline and summary is accounted for by a listed claim.
- No candidate work was dropped: every one the desk found or named was confirmed by its registry.
Code version: fa127b6f988c
What the record does not say
This survey lists what each paper reports; the game has not repeated these results. A search finds papers that match its words, so a listed paper may use AI only in part. The search is limited to the fifty most-cited matches, so the newest results may be missing until they are cited.
How this was written
Landmark works named by the desk, then the most-cited works with a DOI that OpenAlex returns when titles and abstracts are searched for ("protein structure" OR "protein folding" OR "protein design") AND ("machine learning" OR "deep learning" OR "neural network" OR "neural networks" OR "artificial intelligence" OR convolutional OR "reinforcement learning" OR "language model" OR transformer); a hit is kept only if its own title names the field and its title or abstract names an AI method, as whole words, and the hits that name the most of these words in their title come first. Each candidate was looked up at Crossref (OpenAlex as second resolver) and its title, year and first author compared; a work the search found is shown only when Crossref holds it. Landmark summaries were written by the desk; the other works are listed by their own titles.
Source records: code:src/lib/world-grammar/place-grammar.ts#resolvePurpose · doi:10.1038/s41586-021-03819-2 · doi:10.1038/s41467-021-27838-9 · doi:10.1016/s0076-6879(96)66033-9 · doi:10.1126/science.ade2574 · doi:10.1038/s41467-022-32007-7 · doi:10.1186/s13321-018-0285-8 · doi:10.1126/science.abj8754 · doi:10.1038/s41586-019-1923-7 · doi:10.1038/s41586-021-03828-1 · doi:10.1093/nar/gkad1011 · doi:10.1073/pnas.1914677117 · doi:10.1073/pnas.92.19.8700