681:(these relationships are represented by the array shown in red). Internally these refinement transformations contain layers that have the effect of bringing relevant data together and filtering out irrelevant data (the "attention mechanism") for these relationships, in a context-dependent way, learnt from training data. These transformations are iterated, the updated information output by one step becoming the input of the next, with the sharpened residue/residue information feeding into the update of the residue/sequence information, and then the improved residue/sequence information feeding into the update of the residue/residue information. As the iteration progresses, according to one report, the "attention algorithm ... mimics the way a person might assemble a jigsaw puzzle: first connecting pieces in small clumps—in this case clusters of amino acids—and then searching for ways to join the clumps in a larger whole."
36:
506:
807:. but, as stated in the "Read Me" file on that website: "This code can't be used to predict structure of an arbitrary protein sequence. It can be used to predict structure only on the CASP13 dataset (links below). The feature generation code is tightly coupled to our internal infrastructure as well as external tools, hence we are unable to open-source it." Therefore, in essence, the code deposited is not suitable for general use but only for the CASP13 proteins. The company has not announced plans to make their code publicly available as of 5 March 2021.
639:
5152:
4228:
5132:
755:
4238:
940:, the story was widely covered by major national newspapers,. A frequent theme was that ability to predict protein structures accurately based on the constituent amino acid sequence is expected to have a wide variety of benefits in the life sciences space including accelerating advanced drug discovery and enabling better understanding of diseases. Some have noted that even a perfect answer to the protein
631:
1222:, an international open-access database, before releasing the computationally determined structures of the under-studied protein molecules. The team acknowledged that although these protein structures might not be the subject of ongoing therapeutical research efforts, they will add to the community's understanding of the SARS-CoV-2 virus. Specifically, AlphaFold 2's prediction of the structure of the
685:
having a GDT_TS of 78, but with a large number (90%) of stereochemical violations – i.e. unphysical bond angles or lengths. With subsequent iterations the number of stereochemical violations fell. By the third iteration the GDT_TS of the prediction was approaching 90, and by the eighth iteration the number of stereochemical violations was approaching zero.
1674:
Stanislav; Jain, Rishub; Adler, Jonas; Back, Trevor; Petersen, Stig; Reiman, David; Clancy, Ellen; Zielinski, Michal; Steinegger, Martin; Pacholska, Michalina; Berghammer, Tamas; Bodenstein, Sebastian; Silver, David; Vinyals, Oriol; Senior, Andrew W; Kavukcuoglu, Koray; Kohli, Pushmeet; Hassabis, Demis (2021-07-15).
822:(GDT) measure of accuracy, the program achieved a median score of 92.4 (out of 100), meaning that more than half of its predictions were scored at better than 92.4% for having their atoms in more-or-less the right place, a level of accuracy reported to be comparable to experimental techniques like
433:
were available from proteins with a partially similar sequence. A team that used AlphaFold 2 (2020) repeated the placement in the CASP14 competition in
November 2020. The team achieved a level of accuracy much higher than any other group. It scored above 90 for around two-thirds of the proteins in
845:
To additionally verify AlphaFold-2 the conference organisers approached four leading experimental groups for structures they were finding particularly challenging and had been unable to determine. In all four cases the three-dimensional models produced by AlphaFold 2 were sufficiently accurate to
684:
The output of these iterations then informs the final structure prediction module, which also uses transformers, and is itself then iterated. In an example presented by DeepMind, the structure prediction module achieved a correct topology for the target protein on its first iteration, scored as
769:
The orange trend-line shows that by 2020 online prediction servers had been able to learn from and match this performance, while the best other groups (green curve) had on average been able to make some improvements on it. However, the black trend curve shows the degree to which AlphaFold 2 had
1673:
Jumper, John; Evans, Richard; Pritzel, Alexander; Green, Tim; Figurnov, Michael; Ronneberger, Olaf; Tunyasuvunakool, Kathryn; Bates, Russ; Ĺ˝Ădek, Augustin; Potapenko, Anna; Bridgland, Alex; Meyer, Clemens; Kohl, Simon A A; Ballard, Andrew J; Cowie, Andrew; Romera-Paredes, Bernardino; Nikolov,
1132:, but for humans they are available in the whole batch file. AlphaFold planned to add more sequences to the collection, the initial goal (as of beginning of 2022) being to cover most of the UniRef90 set of more than 100 million proteins. As of May 15, 2022, 992,316 predictions were available.
837:
for the set of overlapped C-alpha atoms. 76% of predictions achieved better than 3 Ă…, and 46% had a C-alpha atom RMS accuracy better than 2 Ă…, with a median RMS deviation in its predictions of 2.1 Ă… for a set of overlapped CA atoms. AlphaFold 2 also achieved an accuracy in modelling surface
688:
The training data was originally restricted to single peptide chains. However, the
October 2021 update, named AlphaFold-Multimer, included protein complexes in its training data. DeepMind stated this update succeeded about 70% of the time at accurately predicting protein-protein interactions.
653:
The software design used in AlphaFold 1 contained a number of modules, each trained separately, that were used to produce the guide potential that was then combined with the physics-based energy potential. AlphaFold 2 replaced this with a system of sub-networks coupled together into a single
2232:
Abramson, Josh; Adler, Jonas; Dunger, Jack; Evans, Richard; Green, Tim; Pritzel, Alexander; Ronneberger, Olaf; Willmore, Lindsay; Ballard, Andrew J.; Bambrick, Joshua; Bodenstein, Sebastian W.; Evans, David A.; Hung, Chia-Chun; O’Neill, Michael; Reiman, David (2024-05-08).
826:. In 2018 AlphaFold 1 had only reached this level of accuracy in two of all of its predictions. 88% of predictions in the 2020 competition had a GDT_TS score of more than 80. On the group of targets classed as the most difficult, AlphaFold 2 achieved a median score of 87.
2649:
For the GDT_TS measure used, each atom in the prediction scores a quarter of a point if it is within 8 Ă… (0.80 nm) of the experimental position; half a point if it is within 4 Ă…, three-quarters of a point if it is within 2 Ă…, and a whole point if it is within 1
912:
called the result "a stunning advance on the protein folding problem", adding that "It has occurred decades before many people in the field would have predicted. It will be exciting to see the many ways in which it will fundamentally change biological research."
760:
The crimson trend-line shows how a handful of models including AlphaFold 1 achieved a significant step-change in 2018 over the rate of progress that had previously been achieved, particularly in respect of the protein sequences considered the most difficult to
979:
The open access to source code of several AlphaFold versions (excluding AlphaFold 3) has been provided by DeepMind after requests from the scientific community. Full source code of AlphaFold-3 is expected to be provided to open access by the end of 2024.
1192:
In the algorithm, the residues are moved freely, without any restraints. Therefore, during modeling the integrity of the chain is not maintained. As a result, AlphaFold may produce topologically wrong results, like structures with an arbitrary number of
607:(2018) was built on work developed by various teams in the 2010s, work that looked at the large databanks of related DNA sequences now available from many different organisms (most without known 3D structures), to try to find changes at different
547:, which are all expensive and time-consuming. Such efforts, using the experimental methods, have identified the structures of about 170,000 proteins over the last 60 years, while there are over 200 million known proteins across all life forms.
738:
AlphaFold 3 introduces the "Pairformer", a deep learning architecture inspired from the transformer, considered similar but simpler than the
Evoformer introduced with AlphaFold 2. The raw predictions from the Pairformer module are passed to a
441:
AlphaFold 2's results at CASP14 were described as "astounding" and "transformational". Some researchers noted that the accuracy is not high enough for a third of its predictions, and that it does not reveal the mechanism or rules of
438:(GDT), a test that measures the degree to which a computational program predicted structure is similar to the lab experiment determined structure, with 100 being a complete match, within the distance cutoff used for calculating GDT.
611:
that appeared to be correlated, even though the residues were not consecutive in the main chain. Such correlations suggest that the residues may be close to each other physically, even though not close in the sequence, allowing a
833:(RMS-D) of the placement of the alpha-carbon atoms of the protein backbone chain, which tends to be dominated by the performance of the worst-fitted outliers, 88% of AlphaFold 2's predictions had an RMS deviation of less than 4
873:, a situation AlphaFold was not programmed to consider. For all targets with a single domain, excluding only one very large protein and the two structures determined by NMR, AlphaFold 2 achieved a GDT_TS score of over 80.
1236:. This specific protein is believed to assist the virus in breaking out of the host cell once it replicates. This protein is also believed to play a role in triggering the inflammatory response to the infection.
796:(GDT) score, ahead of 52.5 and 52.4 by the two next best-placed teams, who were also using deep learning to estimate contact distances. Overall, across all targets, the program achieved a GDT score of 68.5.
772:
The detailed spread of data points indicates the degree of consistency or variation achieved by AlphaFold. Outliers represent the handful of sequences for which it did not make such a successful prediction.
654:
differentiable end-to-end model, based entirely on pattern recognition, which was trained in an integrated way as a single integrated structure. Local physics, in the form of energy refinement based on the
1171:
The model relies to some degree upon co-evolutionary information across similar proteins, and thus may not perform well on synthetic proteins or proteins with very low homology to anything in the database.
758:
Results achieved for protein prediction by the best reconstructions in the CASP 2018 competition (small circles) and CASP 2020 competition (large circles), compared with results achieved in previous years.
1189:. Between 50% and 70% of the structures of the human proteome are incomplete without covalently-attached glycans. AlphaFill, a derived database, adds cofactors to AlphaFold models where appropriate.
429:(CASP) in December 2018. The program was particularly successful at predicting the most accurate structure for targets rated as the most difficult by the competition organisers, where no existing
792:
were available from proteins with a partially similar sequence. AlphaFold gave the best prediction for 25 out of 43 protein targets in this class, achieving a median score of 58.9 on the CASP's
3677:
677:
of the protein and another amino acid residue (these relationships are represented by the array shown in green); and between each amino acid position and each different sequences in the input
782:
620:
close the residues might be likely to be—turning the contact map into a likely distance map. It also used more advanced learning methods than previously to develop the inference.
1142:
On July 28, 2022, the team uploaded to the database the structures of around 200 million proteins from 1 million species, covering nearly every known protein on the planet.
2999:
916:
Propelled by press releases from CASP and DeepMind, AlphaFold 2's success received wide media attention. As well as news pieces in the specialist science press, such as
788:
The program was particularly successfully predicting the most accurate structure for targets rated as the most difficult by the competition organisers, where no existing
5026:
2973:
1214:
in the United
Kingdom before release into the larger research community. The team also confirmed accurate prediction against the experimentally determined SARS-CoV-2
590:
identify parts of a larger problem, then piece it together to obtain the overall solution. The overall training was conducted on processing power between 100 and 200
2734:
2915:
2659:
To achieve a GDT_TS score of 92.5, mathematically at least 70% of the structure must be accurate to within 1 Ă…, and at least 85% must be accurate to within 2 Ă…,
578:
DeepMind is known to have trained the program on over 170,000 proteins from a public repository of protein sequences and structures. The program uses a form of
450:
to be considered solved. Nevertheless, there has been widespread respect for the technical achievement. On 15 July 2021 the AlphaFold 2 paper was published in
1840:
295:
858:
2897:
2625:
1426:
4274:
2788:
2319:
1986:
658:
model, is applied only as a final refinement step once the neural network prediction has converged, and only slightly adjusts the predicted structure.
2545:
3624:
2751:
425:
AlphaFold software has had three major versions. A team of researchers that used AlphaFold 1 (2018) placed first in the overall rankings of the 13th
386:
3133:
2769:
4868:
562:
scores of only about 40 out of 100 can be achieved for the most difficult proteins by 2016. AlphaFold started competing in the 2018 CASP using an
1203:
2839:
2880:
2858:
2108:
968:
830:
815:
In
November 2020, DeepMind's new version, AlphaFold 2, won CASP14. Overall, AlphaFold 2 made the best prediction for 88 out of the 97 targets.
2940:
5201:
1475:
1251:"Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13)"
743:, which starts with a cloud of atoms and uses these predictions to iteratively progress towards a 3D depiction of the molecular structure.
3684:
2816:
1451:
426:
182:
147:
551:
2463:
2599:
1210:. The structures of these proteins were pending experimental detection in early 2020. Results were examined by the scientists at the
4029:
3691:
616:
to be estimated. Building on recent work prior to 2018, AlphaFold 1 extended this to estimate a probability distribution for just
558:, which was launched in 1994 to challenge the scientific community to produce their best protein structure predictions, found that
246:
224:
3273:
McBride, John M.; Polev, Konstantin; Abdirasulov, Amirbek; Reinharz, Vladimir; Grzybowski, Bartosz A.; Tlusty, Tsvi (2023-11-20).
1533:
4384:
160:
4267:
2129:
2077:
2051:
1315:
84:
3338:
3053:
5206:
3678:
CASP14: what Google DeepMind's AlphaFold 2 really achieved, and what it means for protein folding, biology and bioinformatics
1844:
1587:
Callaway, Ewen (2020-11-30). "'It will change everything': DeepMind's AI makes gigantic leap in solving protein structures".
379:
305:
259:
214:
209:
1638:
5057:
3589:
662:
500:
3788:
3745:
Mirdita, Milot; SchĂĽtze, Konstantin; Moriwaki, Yoshitaka; Heo, Lim; Ovchinnikov, Sergey; Steinegger, Martin (2022-05-30).
650:, 2020) is significantly different from the original version that won CASP 13 in 2018, according to the team at DeepMind.
5158:
4709:
4446:
4052:
3816:
3157:
1652:
964:
358:
330:
325:
219:
4198:
1229:
318:
187:
177:
167:
5191:
4970:
4597:
4404:
4260:
3722:
1864:
1786:
290:
236:
202:
69:
2200:
2021:
5186:
4925:
1162:
952:
problem—understanding in detail how the folding process actually occurs in nature (and how sometimes they can also
372:
276:
122:
1128:, amounting to over 365,000 proteins. The database does not include proteins with fewer than 16 or more than 2700
2820:
1186:
881:
In 2022 DeepMind did not enter CASP15, but most of the entrants used AlphaFold or tools incorporating AlphaFold.
724:
554:
from their amino acid sequences, but the accuracy of such methods has not been close to experimental techniques.
54:
1816:
5112:
5052:
4650:
1349:
861:
methods, which define protein structure directly in aqueous solution, whereas AlphaFold was mostly trained on
4645:
4334:
2570:
763:(Qualitative improvement had been made in earlier years, but it is only as changes bring structures within 8
5087:
4484:
4441:
4394:
4389:
1114:
942:
789:
579:
496:
430:
415:
2718:
Artificial intelligence solution to a 50-year-old science challenge could 'revolutionise' medical research
5138:
4434:
4360:
4130:
3540:"AlphaFold Blindness to Topological Barriers Affects Its Ability to Correctly Predict Proteins' Topology"
670:
241:
192:
89:
4762:
4697:
4298:
3000:"If Google's Alphafold2 really has solved the protein folding problem, they need to show their working"
2477:"Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13"
857:
Of the three structures that AlphaFold 2 had the least success in predicting, two had been obtained by
544:
532:
of the proteins. The 3-D structure is crucial to understanding the biological function of the protein.
529:
64:
47:
3371:
608:
5163:
5021:
4660:
4491:
4314:
4241:
4187:
1215:
521:
142:
3649:
2717:
1181:
AlphaFold 3 version can predict structures of protein complexes with a very limited set of selected
866:
746:
The AlphaFold server was created to provide free access to AlphaFold 3 for non-commercial research.
35:
5211:
5196:
5062:
4319:
4231:
3923:
2738:
2279:
1233:
902:
540:
266:
5107:
5092:
4745:
4740:
4640:
4508:
4289:
4142:
3809:
1369:
1211:
1182:
563:
447:
403:
27:
3487:
Hekkelman, Maarten L.; de Vries, Ida; Joosten, Robbie P.; Perrakis, Anastassis (February 2023).
2072:
The structure module is stated to use a "3-d equivariant transformer architecture" (John Jumper
711:. AlphaFold 3 is not limited to single-chain proteins, as it can also predict the structures of
5067:
4827:
4546:
4541:
4062:
3917:
3650:"How DeepMind's new protein-folding A.I. is already helping to combat the coronavirus pandemic"
3026:
1303:
1168:
Aphafold-2 was validated for predicting structural effects of mutations with a limited success.
137:
3719:
Open access to protein structure predictions for the human proteome and 20 other key organisms
2859:'Once in a generation advance' as Google AI researchers crack 50-year-old biological challenge
2426:
1563:
5097:
5082:
5047:
4735:
4635:
4503:
4118:
3929:
1304:
Fourteenth
Critical Assessment of Techniques for Protein Structure Prediction (Abstract Book)
1250:
930:
898:
862:
847:
823:
536:
480:
4965:
2804:
1154:
AlphaFold DB provides monomeric models of proteins, rather than their biologically relevant
5117:
5072:
4518:
4463:
4309:
4304:
3983:
3296:
3188:
2488:
2155:
1931:
1687:
1596:
894:
819:
800:
793:
559:
457:
435:
79:
3372:"The case for post-predictional modifications in the AlphaFold Protein Structure Database"
3274:
2974:"La inteligencia artificial arrasa en uno de los problemas más importantes de la biologĂa"
8:
4692:
4670:
4419:
4414:
4372:
4324:
4193:
1129:
231:
3566:
3539:
3300:
3192:
2302:
2265:
2159:
1935:
1691:
1600:
781:
In
December 2018, DeepMind's AlphaFold placed first in the overall rankings of the 13th
5077:
4655:
3802:
3773:
3746:
3515:
3488:
3464:
3432:"Determination of glycosylation sites and site-specific heterogeneity in glycoproteins"
3431:
3412:
3286:
3214:
2700:
2519:
2476:
2408:
2181:
2026:
1960:
1919:
1710:
1675:
1620:
1538:
1493:
1390:
909:
906:
678:
674:
638:
281:
1301:(December 2020), "High Accuracy Protein Structure Prediction Using Deep Learning", in
5143:
5131:
4935:
4587:
4458:
4451:
3882:
3778:
3597:
3571:
3520:
3469:
3451:
3416:
3404:
3396:
3346:
3320:
3312:
3256:
3252:
3218:
3206:
3042:
2948:
2704:
2692:
2524:
2506:
2412:
2400:
2392:
2344:
2252:
2234:
2208:
2185:
2173:
1994:
1965:
1947:
1870:
1715:
1624:
1612:
1497:
1280:
1219:
1207:
1155:
924:
468:
59:
2916:"2023 Breakthrough Prizes Announced: Deepmind's Protein Folders Awarded $ 3 Million"
2105:. It is not known how similar this may or may not be to what was used in AlphaFold.
1860:
1275:
4888:
4878:
4685:
4479:
4429:
4424:
4367:
4355:
4213:
4101:
4089:
3768:
3758:
3561:
3551:
3510:
3500:
3459:
3443:
3386:
3308:
3304:
3196:
2682:
2626:"DeepMind's protein-folding AI has solved a 50-year-old grand challenge of biology"
2514:
2498:
2455:
2382:
2260:
2242:
2163:
1955:
1939:
1883:
1875:
1856:
1705:
1695:
1604:
1483:
1427:"DeepMind's protein-folding AI has solved a 50-year-old grand challenge of biology"
1288:
1261:
1084:
918:
851:
799:
In
January 2020, implementations and illustrative code of AlphaFold 1 was released
767:
of their experimental positions that they start to affect the CASP GDS-TS measure).
452:
197:
132:
117:
2102:
5001:
4945:
4767:
4409:
4329:
4035:
3846:
3260:
2863:
2087:
2083:
1125:
1069:
1024:
948:
740:
712:
704:
525:
443:
74:
3728:
3177:"'The entire protein universe': AI predicts shape of nearly every known protein"
2898:
Protein folding and related problems remain unsolved despite AlphaFold's advance
1787:"AlphaFold 3 predicts the structure and interactions of all of life's molecules"
4975:
4940:
4930:
4755:
4513:
4339:
3947:
3763:
3505:
3447:
3391:
3201:
3176:
3022:
2825:
2687:
2670:
2247:
2168:
2143:
1943:
1887:
1700:
1608:
960:
870:
666:
535:
Protein structures can be determined experimentally through techniques such as
3734:
3556:
3370:
Bagdonas, Haroldas; Fogarty, Carl A.; Fadda, Elisa; Agirre, Jon (2021-10-29).
2387:
2370:
2112:
1488:
1292:
5180:
4920:
4900:
4817:
4496:
4095:
3995:
3601:
3455:
3400:
3350:
3316:
2952:
2774:
2756:
2510:
2396:
2256:
2235:"Accurate structure prediction of biomolecular interactions with AlphaFold 3"
2212:
1998:
1951:
1364:
1334:
936:
708:
583:
567:
419:
411:
127:
3235:
2443:
1161:
Many protein regions are predicted with low confidence score, including the
5006:
4837:
4252:
4208:
3888:
3841:
3782:
3575:
3524:
3473:
3408:
3324:
3210:
2901:
2844:
2696:
2528:
2404:
2280:
Accurate structure prediction of biomolecular interactions with AlphaFold 3
2177:
1969:
1719:
1616:
1344:
1329:
1276:"Improved protein structure prediction using potentials from deep learning"
1175:
550:
Over the years, researchers have applied numerous computational methods to
271:
3625:"Computational predictions of protein structures associated with COVID-19"
5102:
4873:
4782:
4777:
4399:
4377:
4203:
4165:
3740:
3704:
2876:
2291:
1452:"DeepMind solves 50-year-old 'grand challenge' with protein folding A.I."
953:
661:
A key part of the 2020 system are two modules, believed to be based on a
613:
510:
505:
467:
AlphaFold 3 was announced on 8 May 2024. It can predict the structure of
300:
285:
3709:
2881:
The secret of life, part 2: the solution of the protein folding problem.
2821:
London A.I. Lab Claims
Breakthrough That Could Accelerate Drug Discovery
2799:
In all science editor Tom
Whipple wrote six articles on the subject for
1733:
4996:
4955:
4950:
4863:
4772:
4680:
4592:
4572:
4057:
3955:
3089:
2884:
2502:
2459:
2444:
Deep-learning contact-map guided protein structure prediction in CASP13
2094:
1265:
1228:
protein was very similar to the structure determined by researchers at
1109:
was launched on July 22, 2021, as a joint effort between AlphaFold and
839:
2671:"After AlphaFold: protein-folding contest seeks next big breakthrough"
2095:
SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks
4991:
4960:
4858:
4702:
4602:
4556:
4551:
4536:
4148:
3894:
3825:
2793:
2752:
DeepMind's AI biologist can decipher secrets of the machinery of life
1898:
1354:
1089:
901:
and great progress towards a decades-old grand challenge of biology.
335:
99:
3339:"DeepMind's latest AI breakthrough could turbocharge drug discovery"
2789:
Deepmind finds biology's 'holy grail' with answer to protein problem
2770:
The predictions of DeepMind's latest AI could revolutionise medicine
2735:
Protein folding and science communication: Between hype and humility
983:
754:
4893:
4725:
3989:
3430:
An, Hyun Joo; Froehlich, John W; Lebrilla, Carlito B (2009-10-01).
3291:
2600:"AlphaFold: a solution to a 50-year-old grand challenge in biology"
2493:
1136:
1121:
1110:
1049:
1009:
834:
764:
461:
407:
172:
94:
3489:"AlphaFill: enriching AlphaFold models with ligands and cofactors"
3253:
AlphaFold heralds a data-driven revolution in biology and medicine
3114:
3065:
783:
Critical Assessment of Techniques for Protein Structure Prediction
5016:
4853:
4807:
4730:
4630:
4625:
4577:
4023:
3876:
3030:
2144:"What's next for AlphaFold and the AI protein-folding revolution"
2098:
1534:'The game has changed.' AI triumphs at solving protein structures
1359:
1118:
517:
340:
3486:
2546:"DeepMind Breakthrough Helps to Solve How Diseases Invade Cells"
2475:
Hou, Jie; Wu, Tianqi; Cao, Renzhi; Cheng, Jianlin (2019-04-25).
2201:"Google Unveils A.I. for Predicting Behavior of Human Molecules"
5031:
5011:
4883:
4675:
3900:
3836:
3718:
3713:
3685:
AlphaFold2 @ CASP14: "It feels like one's child has left home."
3538:
Dabrowski-Tumanski, Pawel; Stasiak, Andrzej (7 November 2023).
3272:
1902:
1757:
1339:
1139:
has been updated to show AlphaFold predictions when available.
804:
728:
630:
4832:
4812:
4802:
4797:
4792:
4787:
4750:
4582:
4124:
3054:
AlphaFold 3: Stepping into the future of structure prediction
1920:"Highly accurate protein structure prediction with AlphaFold"
1734:"GitHub - deepmind/alphafold: Open source code for AlphaFold"
1676:"Highly accurate protein structure prediction with AlphaFold"
1224:
1202:
AlphaFold has been used to predict structures of proteins of
655:
3794:
2022:
Lessons from DeepMind's breakthrough in protein-folding A.I.
1899:
AlphaFold: Machine learning for protein structure prediction
4822:
4136:
2941:"Team behind AI program AlphaFold win Lasker science prize"
2721:
1987:"DeepMind is answering one of biology's biggest challenges"
890:
591:
555:
3369:
3744:
3537:
2840:
DeepMind AI cracks 50-year-old problem of protein folding
732:
720:
716:
484:
476:
472:
3043:
AlphaFold3 — why did Nature publish it without its code?
2231:
3158:"Putting the power of AlphaFold into the world's hands"
3134:"Alphafold Structure Predictions Available In Interpro"
3027:"Brief update on some exciting progress on #AlphaFold!"
2063:
John Jumper et al., conference abstract (December 2020)
1917:
1672:
587:
1113:. At launch the database contains AlphaFold-predicted
946:
problem would still leave questions about the protein
634:
AlphaFold 2 performance, experiments, and architecture
3747:"ColabFold: Making protein folding accessible to all"
2305:, CASP 13, December 2018. (AlphaFold = Team 043: A7D)
1476:"Structural biology: How proteins got their close-up"
3016:
1981:
1979:
3429:
1871:
A watershed moment for protein structure prediction
1863:. See also Mohammed AlQuraishi (December 9, 2018),
3735:AlphaFold: The making of a scientific breakthrough
3692:The AlphaFold2 Method Paper: A Fount of Good Ideas
2320:"Google's DeepMind predicts 3D shapes of proteins"
1564:CASP14 scores just came out and they're astounding
2594:
2592:
2590:
2481:Proteins: Structure, Function, and Bioinformatics
2448:Proteins: Structure, Function, and Bioinformatics
1976:
1668:
1666:
1558:
1556:
1554:
1552:
1550:
1548:
1528:
1526:
1524:
1522:
1520:
1518:
1516:
1514:
984:Database of protein models generated by AlphaFold
897:(GDT) is considered a significant achievement in
850:. These included target T1100 (Af1503), a small
665:design, which are used to progressively refine a
5178:
3680:, Oxford Protein Informatics Group. (3 December)
3590:"AI Can Help Scientists Find a Covid-19 Vaccine"
3275:"AlphaFold2 Can Predict Single-Mutation Effects"
3230:
3228:
2042:
2040:
2038:
2036:
2313:
2311:
1421:
1419:
1417:
1415:
1413:
1411:
971:for their management of the AlphaFold project.
770:surpassed this again in 2020, across the board.
2587:
2345:"AlphaFold: Using AI for scientific discovery"
2057:
2016:
2014:
1817:"AlphaFold: Using AI for scientific discovery"
1663:
1545:
1511:
969:Albert Lasker Award for Basic Medical Research
869:consisting of 52 identical copies of the same
4268:
3810:
3263:, volume 12, pages 1666–1669, 12 October 2021
3225:
2474:
2118:
2033:
1255:Proteins: Structure, Function, Bioinformatics
1174:The ability of the model to produce multiple
380:
4282:
3259:, Roman A. Laskowski & Neera Borkakoti,
3236:"What use cases does AlphaFold not support?"
2711:
2308:
2303:Group performance based on combined z-scores
1408:
842:described as "really really extraordinary".
3791:for homooligomeric prediction and complexes
2282:, pdf of preprint of the article in Nature.
2011:
1639:No, DeepMind has not solved protein folding
1443:
854:studied by experimentalists for ten years.
456:as an advance access publication alongside
427:Critical Assessment of Structure Prediction
16:Artificial intelligence program by DeepMind
4275:
4261:
3817:
3803:
2082:One design for a transformer network with
1641:, Reciprocal Space (blog), 2 December 2020
846:determine structures of these proteins by
673:" in graph-theory terminology) between an
387:
373:
3772:
3762:
3565:
3555:
3514:
3504:
3463:
3390:
3379:Nature Structural & Molecular Biology
3290:
3200:
2971:
2686:
2518:
2492:
2386:
2264:
2246:
2167:
1959:
1918:Jumper, John; et al. (August 2021).
1913:
1911:
1865:AlphaFold @ CASP13: "What just happened?"
1709:
1699:
1487:
1117:of protein structures of nearly the full
3174:
3131:
2668:
2429:for 043 A7D, 322 Zhang, and 089 MULTICOM
2141:
1586:
1473:
753:
703:was co-developed by Google DeepMind and
637:
629:
504:
1869:Mohammed AlQuraishi (15 January 2020),
1467:
5179:
3725:(AlphaFold Protein Structure Database)
3710:AlphaFold v2.1 code and links to model
3115:"AlphaFold Protein Structure Database"
3090:"AlphaFold Protein Structure Database"
3066:"AlphaFold Protein Structure Database"
2997:
2965:
2938:
2724:organising committee, 30 November 2020
2317:
2292:A non-commercial server of AlphaFold-3
1908:
1758:"AlphaFold Protein Structure Database"
4256:
3798:
3619:
3617:
2540:
2538:
2368:
2046:See block diagram. Also John Jumper
1811:
1809:
1807:
1781:
1779:
1777:
1650:
1449:
1178:conformations of proteins is limited.
692:
623:
597:
586:technique that focuses on having the
552:predict the 3D structures of proteins
460:and a searchable database of species
5113:Generative adversarial network (GAN)
4237:
3438:. Analytical Techniques/Mechanisms.
3363:
3240:AlphaFold Protein Structure Database
2198:
1631:
1582:
1580:
1578:
1576:
1574:
1572:
1474:Stoddart, Charlotte (1 March 2022).
1107:AlphaFold Protein Structure Database
989:AlphaFold Protein Structure Database
889:AlphaFold 2 scoring more than 90 in
642:Architectural details of AlphaFold 2
501:De novo protein structure prediction
5202:Deep learning software applications
4053:Quantum Artificial Intelligence Lab
3436:Current Opinion in Chemical Biology
1150:AlphaFold has various limitations:
965:Breakthrough Prize in Life Sciences
865:. The third exists in nature as a
13:
4199:Generative pre-trained transformer
3670:
3614:
3266:
2535:
1804:
1774:
1651:Balls, Phillip (9 December 2020).
1239:
1230:University of California, Berkeley
530:three dimensional (3-D) structures
34:
14:
5223:
3723:European Bioinformatics Institute
3698:
2913:
1653:"Behind the screens of AlphaFold"
1569:
1318:". Presentation given at CASP 14.
646:The 2020 version of the program (
5151:
5150:
5130:
4236:
4227:
4226:
2142:Callaway, Ewen (13 April 2022).
1187:post-translational modifications
1163:intrinsically disordered protein
1050:https://www.alphafold.ebi.ac.uk/
725:post-translational modifications
416:predictions of protein structure
3642:
3582:
3531:
3480:
3423:
3331:
3246:
3168:
3150:
3125:
3107:
3082:
3058:
3047:
3036:
2991:
2932:
2907:
2890:
2870:
2851:
2832:
2810:
2781:
2763:
2744:
2727:
2662:
2653:
2643:
2618:
2563:
2468:
2432:
2419:
2362:
2337:
2318:Sample, Ian (2 December 2018).
2296:
2285:
2273:
2225:
2192:
2135:
2111:by AlQuaraishi on this, or the
2066:
1892:
1834:
1750:
1726:
1197:
818:On the competition's preferred
749:
418:. The program is designed as a
55:Artificial general intelligence
5063:Recurrent neural network (RNN)
5053:Differentiable neural computer
3309:10.1103/PhysRevLett.131.218401
2972:DomĂnguez, Nuño (2020-12-02).
1644:
1383:
1350:Human Proteome Folding Project
1145:
974:
863:protein structures in crystals
1:
5108:Variational autoencoder (VAE)
5068:Long short-term memory (LSTM)
4335:Computational learning theory
3824:
3175:Callaway, Ewen (2022-07-28).
2669:Callaway, Ewen (2022-12-13).
2371:"Deep learning 3D structures"
2090:was proposed in Fabian Fuchs
1861:10.1093/bioinformatics/btz422
1376:
1135:In July 2021, UniProt-KB and
490:
5207:Molecular modelling software
5088:Convolutional neural network
2998:Briggs, David (2020-12-04).
2803:on the day the news broke. (
2571:"deepmind/deepmind-research"
1566:, Twitter, 30 November 2020.
1004:protein structure prediction
959:In 2023, Demis Hassabis and
884:
573:
509:Amino-acid chains, known as
497:Protein structure prediction
7:
5083:Multilayer perceptron (MLP)
2797:(online), 30 November 2020.
1322:
669:for each relationship (or "
90:Natural language processing
10:
5228:
5159:Artificial neural networks
5073:Gated recurrent unit (GRU)
4299:Differentiable programming
3764:10.1038/s41592-022-01488-1
3506:10.1038/s41592-022-01685-y
3448:10.1016/j.cbpa.2009.07.022
3392:10.1038/s41594-021-00680-9
3202:10.1038/d41586-022-02083-2
2939:Sample, Ian (2023-09-21).
2688:10.1038/d41586-022-04438-1
2248:10.1038/s41586-024-07487-w
2169:10.1038/d41586-022-00997-5
1944:10.1038/s41586-021-03819-2
1888:10.1038/d41586-019-03951-0
1701:10.1038/s41586-021-03819-2
1609:10.1038/d41586-020-03348-4
831:root-mean-square deviation
545:nuclear magnetic resonance
494:
406:(AI) program developed by
143:Hybrid intelligent systems
65:Recursive self-improvement
5126:
5040:
4984:
4913:
4846:
4718:
4618:
4611:
4565:
4529:
4492:Artificial neural network
4472:
4348:
4315:Automatic differentiation
4288:
4222:
4188:Attention Is All You Need
4179:
4158:
4111:
4082:
4075:
4045:
4016:
4009:
3970:
3939:
3910:
3869:
3862:
3855:
3832:
3557:10.3390/molecules28227462
3138:proteinswebteam.github.io
3132:InterPro (22 July 2021).
2388:10.1038/s41592-020-0779-y
2199:Metz, Cade (2024-05-08).
1489:10.1146/knowable-022822-1
1450:Shead, Sam (2020-11-30).
1293:10.1038/s41586-019-1923-7
1206:, the causative agent of
1095:
1083:
1078:
1068:
1063:
1055:
1045:
1040:
1033:
1023:
1018:
1008:
998:
993:
876:
810:
776:
699:Announced on 8 May 2024,
513:, fold to form a protein.
471:created by proteins with
5192:Applied machine learning
4320:Neuromorphic engineering
4283:Differentiable computing
3737:, DeepMind, via YouTube.
2739:University of Nottingham
2487:(12). Wiley: 1165–1178.
2130:AlphaFold 2 presentation
2078:AlphaFold 2 presentation
2052:AlphaFold 2 presentation
1234:cryo-electron microscopy
541:cryo-electron microscopy
267:Artificial consciousness
5187:Bioinformatics software
5093:Residual neural network
4509:Artificial Intelligence
3687:(blog), 8 December 2020
3279:Physical Review Letters
3119:www.alphafold.ebi.ac.uk
2369:Singh, Arunima (2020).
1370:Predicted Aligned Error
1218:that was shared in the
1212:Francis Crick Institute
707:, both subsidiaries of
564:artificial intelligence
448:protein folding problem
404:artificial intelligence
138:Evolutionary algorithms
28:Artificial intelligence
4063:Tensor Processing Unit
3705:AlphaFold-3 web server
773:
643:
635:
514:
39:
5048:Neural Turing machine
4636:Human image synthesis
3690:Mohammed AlQuraishi,
3683:Mohammed AlQuraishi,
2904:blog, 8 December 2020
2741:blog, 4 December 2020
2630:MIT Technology Review
1562:Mohammed AlQuraishi,
1431:MIT Technology Review
1014:all UniProt proteomes
931:MIT Technology Review
899:computational biology
848:molecular replacement
824:X-ray crystallography
757:
667:vector of information
641:
633:
537:X-ray crystallography
522:chains of amino acids
508:
38:
5139:Computer programming
5118:Graph neural network
4693:Text-to-video models
4671:Text-to-image models
4519:Large language model
4504:Scientific computing
4310:Statistical manifold
4305:Information geometry
3694:(blog), 25 July 2021
907:structural biologist
895:global distance test
820:global distance test
794:global distance test
458:open source software
436:global distance test
80:General game playing
4485:In-context learning
4325:Pattern recognition
4194:Future of Go Summit
3301:2023PhRvL.131u8401M
3193:2022Natur.608...15C
3094:alphafold.ebi.ac.uk
3070:alphafold.ebi.ac.uk
2427:CASP 13 data tables
2160:2022Natur.604..234C
2128:(1 December 2020),
2076:(1 December 2020),
2050:(1 December 2020),
1936:2021Natur.596..583J
1845:AlphaFold at CASP13
1841:Mohammed AlQuraishi
1762:alphafold.ebi.ac.uk
1692:2021Natur.596..583J
1601:2020Natur.588..203C
1532:Robert F. Service,
1274:(15 January 2020),
1130:amino acid residues
990:
867:multidomain complex
790:template structures
431:template structures
232:Machine translation
148:Systems integration
85:Knowledge reasoning
22:Part of a series on
5078:Echo state network
4966:JĂĽrgen Schmidhuber
4661:Facial recognition
4656:Speech recognition
4566:Software libraries
3940:In popular culture
2896:e.g. Greg Bowman,
2887:, 30 November 2020
2867:, 30 November 2020
2848:, 30 November 2020
2829:, 30 November 2020
2760:, 30 November 2020
2733:Brigitte Nerlich,
2606:. 30 November 2020
2503:10.1002/prot.25697
2460:10.1002/prot.25792
2205:The New York Times
2113:more detailed post
1542:, 30 November 2020
1314:(December 2020), "
1266:10.1002/prot.25834
988:
910:Venki Ramakrishnan
774:
693:AlphaFold 3 (2024)
679:sequence alignment
675:amino acid residue
644:
636:
624:AlphaFold 2 (2020)
598:AlphaFold 1 (2018)
526:spontaneously fold
515:
410:, a subsidiary of
40:
5174:
5173:
4936:Stephen Grossberg
4909:
4908:
4250:
4249:
4175:
4174:
4071:
4070:
4005:
4004:
3966:
3965:
3856:Computer programs
3676:Carlos Outeiral,
3257:Janet M. Thornton
2778:, 2 December 2020
2750:Michael Le Page,
2720:(press release),
2154:(7905): 234–238.
2132:, slides 12 to 20
2030:, 1 December 2020
1930:(7873): 583–589.
1905:, 31 January 2020
1823:. 15 January 2020
1686:(7873): 583–589.
1595:(7837): 203–204.
1480:Knowable Magazine
1270:Andrew W. Senior
1249:(December 2019),
1245:Andrew W. Senior
1220:Protein Data Bank
1124:of humans and 20
1103:
1102:
713:protein complexes
580:attention network
414:, which performs
397:
396:
133:Bayesian networks
60:Intelligent agent
5219:
5164:Machine learning
5154:
5153:
5134:
4889:Action selection
4879:Self-driving car
4686:Stable Diffusion
4651:Speech synthesis
4616:
4615:
4480:Machine learning
4356:Gradient descent
4277:
4270:
4263:
4254:
4253:
4240:
4239:
4230:
4229:
4214:Google Workspace
4080:
4079:
4014:
4013:
4010:Machine learning
3867:
3866:
3860:
3859:
3819:
3812:
3805:
3796:
3795:
3786:
3776:
3766:
3664:
3663:
3661:
3660:
3646:
3640:
3639:
3637:
3636:
3621:
3612:
3611:
3609:
3608:
3586:
3580:
3579:
3569:
3559:
3535:
3529:
3528:
3518:
3508:
3484:
3478:
3477:
3467:
3427:
3421:
3420:
3394:
3376:
3367:
3361:
3360:
3358:
3357:
3335:
3329:
3328:
3294:
3270:
3264:
3250:
3244:
3243:
3232:
3223:
3222:
3204:
3172:
3166:
3165:
3154:
3148:
3147:
3145:
3144:
3129:
3123:
3122:
3111:
3105:
3104:
3102:
3100:
3086:
3080:
3079:
3077:
3076:
3062:
3056:
3051:
3045:
3040:
3034:
3020:
3014:
3013:
3011:
3010:
2995:
2989:
2988:
2986:
2985:
2969:
2963:
2962:
2960:
2959:
2936:
2930:
2929:
2927:
2926:
2911:
2905:
2894:
2888:
2874:
2868:
2857:Lizzie Roberts,
2855:
2849:
2836:
2830:
2814:
2808:
2785:
2779:
2767:
2761:
2748:
2742:
2731:
2725:
2715:
2709:
2708:
2690:
2666:
2660:
2657:
2651:
2647:
2641:
2640:
2638:
2636:
2622:
2616:
2615:
2613:
2611:
2596:
2585:
2584:
2582:
2581:
2567:
2561:
2560:
2558:
2557:
2542:
2533:
2532:
2522:
2496:
2472:
2466:
2436:
2430:
2423:
2417:
2416:
2390:
2366:
2360:
2359:
2357:
2355:
2341:
2335:
2334:
2332:
2330:
2315:
2306:
2300:
2294:
2289:
2283:
2277:
2271:
2270:
2268:
2250:
2229:
2223:
2222:
2220:
2219:
2196:
2190:
2189:
2171:
2139:
2133:
2122:
2116:
2070:
2064:
2061:
2055:
2044:
2031:
2018:
2009:
2008:
2006:
2005:
1983:
1974:
1973:
1963:
1915:
1906:
1896:
1890:
1855:(22), 4862–4865
1838:
1832:
1831:
1829:
1828:
1813:
1802:
1801:
1799:
1798:
1783:
1772:
1771:
1769:
1768:
1754:
1748:
1747:
1745:
1744:
1730:
1724:
1723:
1713:
1703:
1670:
1661:
1660:
1648:
1642:
1635:
1629:
1628:
1584:
1567:
1560:
1543:
1530:
1509:
1508:
1506:
1504:
1491:
1471:
1465:
1464:
1462:
1461:
1447:
1441:
1440:
1438:
1437:
1423:
1406:
1405:
1403:
1401:
1387:
1307:, pp. 22–24
1034:Primary citation
991:
987:
852:membrane protein
829:Measured by the
389:
382:
375:
296:Existential risk
118:Machine learning
19:
18:
5227:
5226:
5222:
5221:
5220:
5218:
5217:
5216:
5212:Google DeepMind
5197:Protein folding
5177:
5176:
5175:
5170:
5122:
5036:
5002:Google DeepMind
4980:
4946:Geoffrey Hinton
4905:
4842:
4768:Project Debater
4714:
4612:Implementations
4607:
4561:
4525:
4468:
4410:Backpropagation
4344:
4330:Tensor calculus
4284:
4281:
4251:
4246:
4218:
4171:
4154:
4112:Language models
4107:
4067:
4041:
4017:Neural networks
4001:
3962:
3935:
3906:
3851:
3847:Google DeepMind
3828:
3823:
3701:
3673:
3671:Further reading
3668:
3667:
3658:
3656:
3648:
3647:
3643:
3634:
3632:
3631:. 4 August 2020
3623:
3622:
3615:
3606:
3604:
3588:
3587:
3583:
3536:
3532:
3485:
3481:
3428:
3424:
3385:(11): 869–870.
3374:
3368:
3364:
3355:
3353:
3337:
3336:
3332:
3271:
3267:
3261:Nature Medicine
3251:
3247:
3234:
3233:
3226:
3187:(7921): 15–16.
3173:
3169:
3164:. 22 July 2022.
3156:
3155:
3151:
3142:
3140:
3130:
3126:
3113:
3112:
3108:
3098:
3096:
3088:
3087:
3083:
3074:
3072:
3064:
3063:
3059:
3052:
3048:
3041:
3037:
3021:
3017:
3008:
3006:
2996:
2992:
2983:
2981:
2970:
2966:
2957:
2955:
2937:
2933:
2924:
2922:
2912:
2908:
2895:
2891:
2875:
2871:
2864:Daily Telegraph
2856:
2852:
2837:
2833:
2815:
2811:
2798:
2786:
2782:
2768:
2764:
2749:
2745:
2732:
2728:
2716:
2712:
2681:(7942): 13–14.
2667:
2663:
2658:
2654:
2648:
2644:
2634:
2632:
2624:
2623:
2619:
2609:
2607:
2598:
2597:
2588:
2579:
2577:
2569:
2568:
2564:
2555:
2553:
2544:
2543:
2536:
2473:
2469:
2454:(12) 1149–1164
2437:
2433:
2424:
2420:
2367:
2363:
2353:
2351:
2343:
2342:
2338:
2328:
2326:
2316:
2309:
2301:
2297:
2290:
2286:
2278:
2274:
2230:
2226:
2217:
2215:
2197:
2193:
2140:
2136:
2123:
2119:
2115:by Fabian Fuchs
2106:
2081:
2071:
2067:
2062:
2058:
2045:
2034:
2019:
2012:
2003:
2001:
1985:
1984:
1977:
1916:
1909:
1897:
1893:
1868:
1839:
1835:
1826:
1824:
1815:
1814:
1805:
1796:
1794:
1785:
1784:
1775:
1766:
1764:
1756:
1755:
1751:
1742:
1740:
1732:
1731:
1727:
1671:
1664:
1657:Chemistry World
1649:
1645:
1637:Stephen Curry,
1636:
1632:
1585:
1570:
1561:
1546:
1531:
1512:
1502:
1500:
1472:
1468:
1459:
1457:
1448:
1444:
1435:
1433:
1425:
1424:
1409:
1399:
1397:
1389:
1388:
1384:
1379:
1374:
1325:
1260:(12) 1141–1148
1242:
1240:Published works
1200:
1148:
1126:model organisms
1096:Curation policy
1025:Research center
1000:
986:
977:
967:as well as the
887:
879:
813:
779:
771:
768:
762:
759:
752:
741:diffusion model
705:Isomorphic Labs
695:
626:
600:
576:
503:
493:
444:protein folding
393:
364:
363:
354:
346:
345:
321:
311:
310:
282:Control problem
262:
252:
251:
163:
153:
152:
113:
105:
104:
75:Computer vision
50:
17:
12:
11:
5:
5225:
5215:
5214:
5209:
5204:
5199:
5194:
5189:
5172:
5171:
5169:
5168:
5167:
5166:
5161:
5148:
5147:
5146:
5141:
5127:
5124:
5123:
5121:
5120:
5115:
5110:
5105:
5100:
5095:
5090:
5085:
5080:
5075:
5070:
5065:
5060:
5055:
5050:
5044:
5042:
5038:
5037:
5035:
5034:
5029:
5024:
5019:
5014:
5009:
5004:
4999:
4994:
4988:
4986:
4982:
4981:
4979:
4978:
4976:Ilya Sutskever
4973:
4968:
4963:
4958:
4953:
4948:
4943:
4941:Demis Hassabis
4938:
4933:
4931:Ian Goodfellow
4928:
4923:
4917:
4915:
4911:
4910:
4907:
4906:
4904:
4903:
4898:
4897:
4896:
4886:
4881:
4876:
4871:
4866:
4861:
4856:
4850:
4848:
4844:
4843:
4841:
4840:
4835:
4830:
4825:
4820:
4815:
4810:
4805:
4800:
4795:
4790:
4785:
4780:
4775:
4770:
4765:
4760:
4759:
4758:
4748:
4743:
4738:
4733:
4728:
4722:
4720:
4716:
4715:
4713:
4712:
4707:
4706:
4705:
4700:
4690:
4689:
4688:
4683:
4678:
4668:
4663:
4658:
4653:
4648:
4643:
4638:
4633:
4628:
4622:
4620:
4613:
4609:
4608:
4606:
4605:
4600:
4595:
4590:
4585:
4580:
4575:
4569:
4567:
4563:
4562:
4560:
4559:
4554:
4549:
4544:
4539:
4533:
4531:
4527:
4526:
4524:
4523:
4522:
4521:
4514:Language model
4511:
4506:
4501:
4500:
4499:
4489:
4488:
4487:
4476:
4474:
4470:
4469:
4467:
4466:
4464:Autoregression
4461:
4456:
4455:
4454:
4444:
4442:Regularization
4439:
4438:
4437:
4432:
4427:
4417:
4412:
4407:
4405:Loss functions
4402:
4397:
4392:
4387:
4382:
4381:
4380:
4370:
4365:
4364:
4363:
4352:
4350:
4346:
4345:
4343:
4342:
4340:Inductive bias
4337:
4332:
4327:
4322:
4317:
4312:
4307:
4302:
4294:
4292:
4286:
4285:
4280:
4279:
4272:
4265:
4257:
4248:
4247:
4245:
4244:
4234:
4223:
4220:
4219:
4217:
4216:
4211:
4206:
4201:
4196:
4191:
4183:
4181:
4177:
4176:
4173:
4172:
4170:
4169:
4162:
4160:
4156:
4155:
4153:
4152:
4146:
4140:
4134:
4128:
4122:
4115:
4113:
4109:
4108:
4106:
4105:
4099:
4093:
4086:
4084:
4077:
4073:
4072:
4069:
4068:
4066:
4065:
4060:
4055:
4049:
4047:
4043:
4042:
4040:
4039:
4033:
4027:
4020:
4018:
4011:
4007:
4006:
4003:
4002:
4000:
3999:
3993:
3987:
3981:
3974:
3972:
3968:
3967:
3964:
3963:
3961:
3960:
3952:
3943:
3941:
3937:
3936:
3934:
3933:
3927:
3921:
3914:
3912:
3908:
3907:
3905:
3904:
3898:
3892:
3886:
3880:
3873:
3871:
3864:
3857:
3853:
3852:
3850:
3849:
3844:
3839:
3833:
3830:
3829:
3822:
3821:
3814:
3807:
3799:
3793:
3792:
3757:(6): 679–682.
3751:Nature Methods
3738:
3732:
3726:
3716:
3707:
3700:
3699:External links
3697:
3696:
3695:
3688:
3681:
3672:
3669:
3666:
3665:
3641:
3613:
3581:
3530:
3499:(2): 205–213.
3493:Nature Methods
3479:
3442:(4): 421–426.
3422:
3362:
3330:
3285:(21): 218401.
3265:
3245:
3224:
3167:
3149:
3124:
3106:
3081:
3057:
3046:
3035:
3033:, 18 June 2021
3023:Demis Hassabis
3015:
2990:
2964:
2931:
2906:
2889:
2869:
2850:
2831:
2826:New York Times
2809:
2780:
2762:
2743:
2726:
2710:
2661:
2652:
2642:
2617:
2586:
2562:
2534:
2494:10.1101/552422
2467:
2431:
2418:
2375:Nature Methods
2361:
2336:
2307:
2295:
2284:
2272:
2224:
2191:
2134:
2117:
2065:
2056:
2032:
2010:
1993:. 2020-11-30.
1975:
1907:
1891:
1849:Bioinformatics
1833:
1803:
1773:
1749:
1725:
1662:
1643:
1630:
1568:
1544:
1510:
1466:
1442:
1407:
1381:
1380:
1378:
1375:
1373:
1372:
1367:
1362:
1357:
1352:
1347:
1342:
1337:
1332:
1326:
1324:
1321:
1320:
1319:
1308:
1295:
1268:
1241:
1238:
1199:
1196:
1195:
1194:
1190:
1179:
1172:
1169:
1166:
1159:
1147:
1144:
1101:
1100:
1097:
1093:
1092:
1087:
1081:
1080:
1076:
1075:
1072:
1066:
1065:
1061:
1060:
1057:
1053:
1052:
1047:
1043:
1042:
1038:
1037:
1035:
1031:
1030:
1027:
1021:
1020:
1016:
1015:
1012:
1006:
1005:
1002:
996:
995:
985:
982:
976:
973:
886:
883:
878:
875:
812:
809:
778:
775:
751:
748:
694:
691:
625:
622:
599:
596:
575:
572:
492:
489:
395:
394:
392:
391:
384:
377:
369:
366:
365:
362:
361:
355:
352:
351:
348:
347:
344:
343:
338:
333:
328:
322:
317:
316:
313:
312:
309:
308:
303:
298:
293:
288:
279:
274:
269:
263:
258:
257:
254:
253:
250:
249:
244:
239:
234:
229:
228:
227:
217:
212:
207:
206:
205:
200:
195:
185:
180:
178:Earth sciences
175:
170:
168:Bioinformatics
164:
159:
158:
155:
154:
151:
150:
145:
140:
135:
130:
125:
120:
114:
111:
110:
107:
106:
103:
102:
97:
92:
87:
82:
77:
72:
67:
62:
57:
51:
46:
45:
42:
41:
31:
30:
24:
23:
15:
9:
6:
4:
3:
2:
5224:
5213:
5210:
5208:
5205:
5203:
5200:
5198:
5195:
5193:
5190:
5188:
5185:
5184:
5182:
5165:
5162:
5160:
5157:
5156:
5149:
5145:
5142:
5140:
5137:
5136:
5133:
5129:
5128:
5125:
5119:
5116:
5114:
5111:
5109:
5106:
5104:
5101:
5099:
5096:
5094:
5091:
5089:
5086:
5084:
5081:
5079:
5076:
5074:
5071:
5069:
5066:
5064:
5061:
5059:
5056:
5054:
5051:
5049:
5046:
5045:
5043:
5041:Architectures
5039:
5033:
5030:
5028:
5025:
5023:
5020:
5018:
5015:
5013:
5010:
5008:
5005:
5003:
5000:
4998:
4995:
4993:
4990:
4989:
4987:
4985:Organizations
4983:
4977:
4974:
4972:
4969:
4967:
4964:
4962:
4959:
4957:
4954:
4952:
4949:
4947:
4944:
4942:
4939:
4937:
4934:
4932:
4929:
4927:
4924:
4922:
4921:Yoshua Bengio
4919:
4918:
4916:
4912:
4902:
4901:Robot control
4899:
4895:
4892:
4891:
4890:
4887:
4885:
4882:
4880:
4877:
4875:
4872:
4870:
4867:
4865:
4862:
4860:
4857:
4855:
4852:
4851:
4849:
4845:
4839:
4836:
4834:
4831:
4829:
4826:
4824:
4821:
4819:
4818:Chinchilla AI
4816:
4814:
4811:
4809:
4806:
4804:
4801:
4799:
4796:
4794:
4791:
4789:
4786:
4784:
4781:
4779:
4776:
4774:
4771:
4769:
4766:
4764:
4761:
4757:
4754:
4753:
4752:
4749:
4747:
4744:
4742:
4739:
4737:
4734:
4732:
4729:
4727:
4724:
4723:
4721:
4717:
4711:
4708:
4704:
4701:
4699:
4696:
4695:
4694:
4691:
4687:
4684:
4682:
4679:
4677:
4674:
4673:
4672:
4669:
4667:
4664:
4662:
4659:
4657:
4654:
4652:
4649:
4647:
4644:
4642:
4639:
4637:
4634:
4632:
4629:
4627:
4624:
4623:
4621:
4617:
4614:
4610:
4604:
4601:
4599:
4596:
4594:
4591:
4589:
4586:
4584:
4581:
4579:
4576:
4574:
4571:
4570:
4568:
4564:
4558:
4555:
4553:
4550:
4548:
4545:
4543:
4540:
4538:
4535:
4534:
4532:
4528:
4520:
4517:
4516:
4515:
4512:
4510:
4507:
4505:
4502:
4498:
4497:Deep learning
4495:
4494:
4493:
4490:
4486:
4483:
4482:
4481:
4478:
4477:
4475:
4471:
4465:
4462:
4460:
4457:
4453:
4450:
4449:
4448:
4445:
4443:
4440:
4436:
4433:
4431:
4428:
4426:
4423:
4422:
4421:
4418:
4416:
4413:
4411:
4408:
4406:
4403:
4401:
4398:
4396:
4393:
4391:
4388:
4386:
4385:Hallucination
4383:
4379:
4376:
4375:
4374:
4371:
4369:
4366:
4362:
4359:
4358:
4357:
4354:
4353:
4351:
4347:
4341:
4338:
4336:
4333:
4331:
4328:
4326:
4323:
4321:
4318:
4316:
4313:
4311:
4308:
4306:
4303:
4301:
4300:
4296:
4295:
4293:
4291:
4287:
4278:
4273:
4271:
4266:
4264:
4259:
4258:
4255:
4243:
4235:
4233:
4225:
4224:
4221:
4215:
4212:
4210:
4207:
4205:
4202:
4200:
4197:
4195:
4192:
4189:
4185:
4184:
4182:
4178:
4167:
4164:
4163:
4161:
4157:
4150:
4147:
4144:
4141:
4138:
4135:
4132:
4129:
4126:
4123:
4120:
4117:
4116:
4114:
4110:
4103:
4100:
4097:
4094:
4091:
4088:
4087:
4085:
4081:
4078:
4076:Generative AI
4074:
4064:
4061:
4059:
4056:
4054:
4051:
4050:
4048:
4044:
4037:
4034:
4031:
4028:
4025:
4022:
4021:
4019:
4015:
4012:
4008:
3997:
3996:AlphaGeometry
3994:
3991:
3988:
3985:
3982:
3979:
3976:
3975:
3973:
3969:
3958:
3957:
3953:
3950:
3949:
3945:
3944:
3942:
3938:
3931:
3928:
3925:
3922:
3919:
3916:
3915:
3913:
3909:
3902:
3899:
3896:
3893:
3890:
3887:
3884:
3881:
3878:
3875:
3874:
3872:
3868:
3865:
3861:
3858:
3854:
3848:
3845:
3843:
3840:
3838:
3835:
3834:
3831:
3827:
3820:
3815:
3813:
3808:
3806:
3801:
3800:
3797:
3790:
3784:
3780:
3775:
3770:
3765:
3760:
3756:
3752:
3748:
3742:
3739:
3736:
3733:
3730:
3727:
3724:
3720:
3717:
3715:
3711:
3708:
3706:
3703:
3702:
3693:
3689:
3686:
3682:
3679:
3675:
3674:
3655:
3651:
3645:
3630:
3626:
3620:
3618:
3603:
3599:
3595:
3591:
3585:
3577:
3573:
3568:
3563:
3558:
3553:
3549:
3545:
3541:
3534:
3526:
3522:
3517:
3512:
3507:
3502:
3498:
3494:
3490:
3483:
3475:
3471:
3466:
3461:
3457:
3453:
3449:
3445:
3441:
3437:
3433:
3426:
3418:
3414:
3410:
3406:
3402:
3398:
3393:
3388:
3384:
3380:
3373:
3366:
3352:
3348:
3344:
3340:
3334:
3326:
3322:
3318:
3314:
3310:
3306:
3302:
3298:
3293:
3288:
3284:
3280:
3276:
3269:
3262:
3258:
3254:
3249:
3241:
3237:
3231:
3229:
3220:
3216:
3212:
3208:
3203:
3198:
3194:
3190:
3186:
3182:
3178:
3171:
3163:
3159:
3153:
3139:
3135:
3128:
3120:
3116:
3110:
3095:
3091:
3085:
3071:
3067:
3061:
3055:
3050:
3044:
3039:
3032:
3029:(tweet), via
3028:
3024:
3019:
3005:
3001:
2994:
2979:
2975:
2968:
2954:
2950:
2946:
2942:
2935:
2921:
2917:
2914:Knapp, Alex.
2910:
2903:
2899:
2893:
2886:
2882:
2878:
2873:
2866:
2865:
2860:
2854:
2847:
2846:
2841:
2835:
2828:
2827:
2822:
2818:
2813:
2806:
2802:
2796:
2795:
2790:
2787:Tom Whipple,
2784:
2777:
2776:
2775:New Scientist
2771:
2766:
2759:
2758:
2757:New Scientist
2753:
2747:
2740:
2736:
2730:
2723:
2719:
2714:
2706:
2702:
2698:
2694:
2689:
2684:
2680:
2676:
2672:
2665:
2656:
2646:
2631:
2627:
2621:
2605:
2601:
2595:
2593:
2591:
2576:
2572:
2566:
2551:
2550:Bloomberg.com
2547:
2541:
2539:
2530:
2526:
2521:
2516:
2512:
2508:
2504:
2500:
2495:
2490:
2486:
2482:
2478:
2471:
2465:
2461:
2457:
2453:
2449:
2445:
2441:
2435:
2428:
2422:
2414:
2410:
2406:
2402:
2398:
2394:
2389:
2384:
2380:
2376:
2372:
2365:
2350:
2346:
2340:
2325:
2321:
2314:
2312:
2304:
2299:
2293:
2288:
2281:
2276:
2267:
2262:
2258:
2254:
2249:
2244:
2240:
2236:
2228:
2214:
2210:
2206:
2202:
2195:
2187:
2183:
2179:
2175:
2170:
2165:
2161:
2157:
2153:
2149:
2145:
2138:
2131:
2127:
2121:
2114:
2110:
2109:the blog post
2104:
2100:
2096:
2093:
2089:
2085:
2079:
2075:
2069:
2060:
2053:
2049:
2043:
2041:
2039:
2037:
2029:
2028:
2023:
2020:Jeremy Kahn,
2017:
2015:
2000:
1996:
1992:
1991:The Economist
1988:
1982:
1980:
1971:
1967:
1962:
1957:
1953:
1949:
1945:
1941:
1937:
1933:
1929:
1925:
1921:
1914:
1912:
1904:
1900:
1895:
1889:
1885:
1881:
1878:
1877:
1872:
1866:
1862:
1858:
1854:
1850:
1846:
1842:
1837:
1822:
1818:
1812:
1810:
1808:
1792:
1788:
1782:
1780:
1778:
1763:
1759:
1753:
1739:
1735:
1729:
1721:
1717:
1712:
1707:
1702:
1697:
1693:
1689:
1685:
1681:
1677:
1669:
1667:
1658:
1654:
1647:
1640:
1634:
1626:
1622:
1618:
1614:
1610:
1606:
1602:
1598:
1594:
1590:
1583:
1581:
1579:
1577:
1575:
1573:
1565:
1559:
1557:
1555:
1553:
1551:
1549:
1541:
1540:
1535:
1529:
1527:
1525:
1523:
1521:
1519:
1517:
1515:
1499:
1495:
1490:
1485:
1481:
1477:
1470:
1456:
1453:
1446:
1432:
1428:
1422:
1420:
1418:
1416:
1414:
1412:
1396:
1392:
1386:
1382:
1371:
1368:
1366:
1365:AlphaGeometry
1363:
1361:
1358:
1356:
1353:
1351:
1348:
1346:
1343:
1341:
1338:
1336:
1335:IBM Blue Gene
1333:
1331:
1328:
1327:
1317:
1313:
1309:
1306:
1305:
1300:
1296:
1294:
1290:
1286:
1283:
1282:
1277:
1273:
1269:
1267:
1263:
1259:
1256:
1252:
1248:
1244:
1243:
1237:
1235:
1231:
1227:
1226:
1221:
1217:
1216:spike protein
1213:
1209:
1205:
1191:
1188:
1184:
1180:
1177:
1173:
1170:
1167:
1164:
1160:
1157:
1153:
1152:
1151:
1143:
1140:
1138:
1133:
1131:
1127:
1123:
1120:
1116:
1112:
1108:
1098:
1094:
1091:
1088:
1086:
1082:
1079:Miscellaneous
1077:
1073:
1071:
1067:
1062:
1058:
1054:
1051:
1048:
1044:
1039:
1036:
1032:
1028:
1026:
1022:
1017:
1013:
1011:
1007:
1003:
997:
992:
981:
972:
970:
966:
962:
957:
955:
951:
950:
945:
944:
939:
938:
937:New Scientist
933:
932:
927:
926:
921:
920:
914:
911:
908:
904:
900:
896:
892:
882:
874:
872:
868:
864:
860:
855:
853:
849:
843:
841:
836:
832:
827:
825:
821:
816:
808:
806:
802:
797:
795:
791:
786:
784:
766:
756:
747:
744:
742:
736:
734:
730:
727:and selected
726:
722:
718:
714:
710:
706:
702:
697:
690:
686:
682:
680:
676:
672:
668:
664:
659:
657:
651:
649:
640:
632:
628:
621:
619:
615:
610:
606:
602:
595:
593:
589:
585:
584:deep learning
581:
571:
569:
568:deep learning
565:
561:
557:
553:
548:
546:
542:
538:
533:
531:
527:
523:
519:
512:
507:
502:
498:
488:
486:
482:
478:
474:
470:
465:
463:
459:
455:
454:
449:
445:
439:
437:
432:
428:
423:
421:
420:deep learning
417:
413:
409:
405:
401:
390:
385:
383:
378:
376:
371:
370:
368:
367:
360:
357:
356:
350:
349:
342:
339:
337:
334:
332:
329:
327:
324:
323:
320:
315:
314:
307:
304:
302:
299:
297:
294:
292:
289:
287:
283:
280:
278:
275:
273:
270:
268:
265:
264:
261:
256:
255:
248:
245:
243:
240:
238:
235:
233:
230:
226:
225:Mental health
223:
222:
221:
218:
216:
213:
211:
208:
204:
201:
199:
196:
194:
191:
190:
189:
188:Generative AI
186:
184:
181:
179:
176:
174:
171:
169:
166:
165:
162:
157:
156:
149:
146:
144:
141:
139:
136:
134:
131:
129:
128:Deep learning
126:
124:
121:
119:
116:
115:
109:
108:
101:
98:
96:
93:
91:
88:
86:
83:
81:
78:
76:
73:
71:
68:
66:
63:
61:
58:
56:
53:
52:
49:
44:
43:
37:
33:
32:
29:
26:
25:
21:
20:
5007:Hugging Face
4971:David Silver
4665:
4619:Audio–visual
4473:Applications
4452:Augmentation
4297:
4209:Google Pixel
3977:
3954:
3946:
3911:Competitions
3889:AlphaGo Zero
3842:Google Brain
3754:
3750:
3657:. Retrieved
3653:
3644:
3633:. Retrieved
3628:
3605:. Retrieved
3593:
3584:
3550:(22): 7462.
3547:
3543:
3533:
3496:
3492:
3482:
3439:
3435:
3425:
3382:
3378:
3365:
3354:. Retrieved
3343:Fast Company
3342:
3333:
3282:
3278:
3268:
3248:
3239:
3184:
3180:
3170:
3161:
3152:
3141:. Retrieved
3137:
3127:
3118:
3109:
3097:. Retrieved
3093:
3084:
3073:. Retrieved
3069:
3060:
3049:
3038:
3018:
3007:. Retrieved
3003:
2993:
2982:. Retrieved
2980:(in Spanish)
2977:
2967:
2956:. Retrieved
2945:The Guardian
2944:
2934:
2923:. Retrieved
2919:
2909:
2902:Folding@home
2892:
2872:
2862:
2853:
2845:The Guardian
2843:
2834:
2824:
2812:
2800:
2792:
2783:
2773:
2765:
2755:
2746:
2729:
2713:
2678:
2674:
2664:
2655:
2645:
2633:. Retrieved
2629:
2620:
2608:. Retrieved
2603:
2578:. Retrieved
2574:
2565:
2554:. Retrieved
2552:. 2020-11-30
2549:
2484:
2480:
2470:
2451:
2447:
2439:
2434:
2421:
2378:
2374:
2364:
2352:. Retrieved
2348:
2339:
2327:. Retrieved
2324:The Guardian
2323:
2298:
2287:
2275:
2238:
2227:
2216:. Retrieved
2204:
2194:
2151:
2147:
2137:
2125:
2124:John Jumper
2120:
2091:
2088:equivariance
2080:, slide 12).
2073:
2068:
2059:
2047:
2025:
2002:. Retrieved
1990:
1927:
1923:
1894:
1879:
1874:
1867:(blog post).
1852:
1848:
1843:(May 2019),
1836:
1825:. Retrieved
1820:
1795:. Retrieved
1793:. 2024-05-08
1790:
1765:. Retrieved
1761:
1752:
1741:. Retrieved
1737:
1728:
1683:
1679:
1656:
1646:
1633:
1592:
1588:
1537:
1501:. Retrieved
1479:
1469:
1458:. Retrieved
1454:
1445:
1434:. Retrieved
1430:
1398:. Retrieved
1394:
1385:
1345:Rosetta@home
1330:Folding@home
1311:
1310:John Jumper
1302:
1298:
1297:John Jumper
1284:
1279:
1271:
1257:
1254:
1246:
1223:
1201:
1198:Applications
1185:and co- and
1149:
1141:
1134:
1106:
1104:
1056:Download URL
978:
958:
947:
941:
935:
929:
923:
917:
915:
888:
880:
856:
844:
828:
817:
814:
798:
787:
780:
750:Competitions
745:
737:
700:
698:
696:
687:
683:
660:
652:
647:
645:
627:
617:
604:
603:
601:
577:
549:
534:
528:to form the
516:
511:polypeptides
466:
451:
440:
424:
399:
398:
272:Chinese room
161:Applications
5155:Categories
5103:Autoencoder
5058:Transformer
4926:Alex Graves
4874:OpenAI Five
4778:IBM Watsonx
4400:Convolution
4378:Overfitting
4204:Google Labs
4030:Transformer
3004:The Skeptic
2877:Tim Hubbard
2838:Ian Sample,
2635:30 November
2610:30 November
2354:30 November
2329:30 November
2101:2020; also
1400:30 November
1391:"AlphaFold"
1316:AlphaFold 2
1146:Limitations
975:Source code
961:John Jumper
905:winner and
903:Nobel Prize
859:protein NMR
840:side chains
801:open-source
701:AlphaFold 3
663:transformer
648:AlphaFold 2
614:contact map
605:AlphaFold 1
570:technique.
520:consist of
301:Turing test
277:Friendly AI
48:Major goals
5181:Categories
5144:Technology
4997:EleutherAI
4956:Fei-Fei Li
4951:Yann LeCun
4864:Q-learning
4847:Decisional
4773:IBM Watson
4681:Midjourney
4573:TensorFlow
4420:Activation
4373:Regression
4368:Clustering
4131:Chinchilla
4058:TensorFlow
3956:The MANIAC
3659:2020-12-01
3635:2020-12-01
3607:2020-12-01
3356:2023-01-24
3292:2204.06860
3143:2021-07-29
3075:2021-07-29
3009:2024-05-12
2984:2024-05-12
2958:2024-05-09
2925:2024-05-09
2885:medium.com
2580:2020-11-30
2556:2020-11-30
2438:Wei Zheng
2381:(3): 249.
2218:2024-05-09
2054:, slide 10
2004:2020-11-30
1882:, 627–628
1827:2020-11-30
1797:2024-05-09
1767:2021-07-24
1743:2021-07-24
1460:2020-11-30
1436:2020-11-30
1377:References
1204:SARS-CoV-2
999:Data types
943:prediction
495:See also:
491:Background
479:, various
306:Regulation
260:Philosophy
215:Healthcare
210:Government
112:Approaches
5027:MIT CSAIL
4992:Anthropic
4961:Andrew Ng
4859:AlphaZero
4703:VideoPoet
4666:AlphaFold
4603:MindSpore
4557:SpiNNaker
4552:Memristor
4459:Diffusion
4435:Rectifier
4415:Batchnorm
4395:Attention
4390:Adversary
4149:VideoPoet
4090:Assistant
3984:AlphaStar
3978:AlphaFold
3924:Lee Sedol
3895:AlphaZero
3826:Google AI
3741:ColabFold
3602:1059-1028
3544:Molecules
3456:1367-5931
3417:240228913
3401:1545-9985
3351:1085-9241
3317:0031-9007
3219:251159714
2953:0261-3077
2817:Cade Metz
2801:The Times
2794:The Times
2705:254660427
2511:0887-3585
2413:212403708
2397:1548-7105
2257:1476-4687
2213:0362-4331
2186:248156195
2107:See also
1999:0013-0613
1952:1476-4687
1625:227243204
1498:247206999
1355:AlphaZero
1183:cofactors
1156:complexes
1099:automatic
1090:CC-BY 4.0
1010:Organisms
885:Reception
574:Algorithm
469:complexes
462:proteomes
400:AlphaFold
336:AI winter
237:Military
100:AI safety
5135:Portals
4894:Auto-GPT
4726:Word2vec
4530:Hardware
4447:Datasets
4349:Concepts
4232:Category
4180:See also
4083:Chatbots
3990:AlphaDev
3870:Versions
3783:35637307
3629:Deepmind
3576:38005184
3567:10672856
3525:36424442
3474:19700364
3409:34716446
3325:38072605
3211:35902752
3162:Deepmind
2697:36513827
2604:Deepmind
2529:30985027
2405:32132733
2349:Deepmind
2266:11168924
2178:35418629
1970:34265844
1821:Deepmind
1720:34265844
1617:33257889
1503:25 March
1395:Deepmind
1323:See also
1287:706–710
1208:COVID-19
1165:regions.
1137:InterPro
1122:proteome
1111:EMBL-EBI
1029:EMBL-EBI
1001:captured
963:won the
785:(CASP).
761:predict.
709:Alphabet
609:residues
518:Proteins
446:for the
422:system.
412:Alphabet
408:DeepMind
359:Glossary
353:Glossary
331:Progress
326:Timeline
286:Takeover
247:Projects
220:Industry
183:Finance
173:Deepfake
123:Symbolic
95:Robotics
70:Planning
5017:Meta AI
4854:AlphaGo
4838:PanGu-ÎŁ
4808:ChatGPT
4783:Granite
4731:Seq2seq
4710:Whisper
4631:WaveNet
4626:AlexNet
4598:Flux.jl
4578:PyTorch
4430:Sigmoid
4425:Softmax
4290:General
4242:Commons
4096:Sparrow
4024:WaveNet
3948:AlphaGo
3918:Fan Hui
3877:AlphaGo
3863:AlphaGo
3789:version
3774:9184281
3731:website
3729:CASP 14
3654:Fortune
3516:9911346
3465:2749913
3297:Bibcode
3189:Bibcode
3099:27 July
3031:twitter
2978:El PaĂs
2520:6800999
2489:bioRxiv
2241:: 1–3.
2156:Bibcode
2103:website
2099:NeurIPS
2027:Fortune
1961:8371605
1932:Bibcode
1711:8371605
1688:Bibcode
1597:Bibcode
1539:Science
1360:AlphaGo
1119:UniProt
1085:License
1046:Website
1019:Contact
994:Content
954:misfold
949:folding
925:Science
729:ligands
481:ligands
434:CASP's
341:AI boom
319:History
242:Physics
5032:Huawei
5012:OpenAI
4914:People
4884:MuZero
4746:Gemini
4741:Claude
4676:DALL-E
4588:Theano
4168:(2024)
4151:(2024)
4145:(2023)
4143:Gemini
4139:(2022)
4133:(2022)
4127:(2021)
4121:(2018)
4104:(2023)
4102:Gemini
4098:(2022)
4092:(2016)
4038:(2022)
4032:(2017)
4026:(2016)
3998:(2024)
3992:(2023)
3986:(2019)
3980:(2018)
3959:(2023)
3951:(2017)
3932:(2017)
3930:Ke Jie
3926:(2016)
3920:(2015)
3903:(2019)
3901:MuZero
3897:(2017)
3891:(2017)
3885:(2016)
3883:Master
3879:(2015)
3837:Google
3781:
3771:
3714:GitHub
3600:
3574:
3564:
3523:
3513:
3472:
3462:
3454:
3415:
3407:
3399:
3349:
3323:
3315:
3217:
3209:
3181:Nature
2951:
2920:Forbes
2805:thread
2703:
2695:
2675:Nature
2575:GitHub
2527:
2517:
2509:
2491:
2464:slides
2462:; and
2411:
2403:
2395:
2263:
2255:
2239:Nature
2211:
2184:
2176:
2148:Nature
2126:et al.
2074:et al.
2048:et al.
1997:
1968:
1958:
1950:
1924:Nature
1903:Foldit
1876:Nature
1791:Google
1738:GitHub
1718:
1708:
1680:Nature
1623:
1615:
1589:Nature
1496:
1340:Foldit
1312:et al.
1299:et al.
1281:Nature
1272:et al.
1247:et al.
1232:using
1193:knots.
1176:native
1115:models
1041:Access
934:, and
919:Nature
877:CASP15
871:domain
811:CASP14
805:GitHub
777:CASP13
524:which
483:, and
453:Nature
402:is an
291:Ethics
5098:Mamba
4869:SARSA
4833:LLaMA
4828:BLOOM
4813:GPT-J
4803:GPT-4
4798:GPT-3
4793:GPT-2
4788:GPT-1
4751:LaMDA
4583:Keras
4159:Other
4125:LaMDA
4046:Other
3971:Other
3594:Wired
3413:S2CID
3375:(PDF)
3287:arXiv
3255:, by
3215:S2CID
2701:S2CID
2440:et al
2409:S2CID
2182:S2CID
2092:et al
2084:SE(3)
1621:S2CID
1494:S2CID
1225:ORF3a
1064:Tools
715:with
656:AMBER
566:(AI)
203:Music
198:Audio
5022:Mila
4823:PaLM
4756:Bard
4736:BERT
4719:Text
4698:Sora
4166:Vids
4137:PaLM
4119:BERT
4036:Gato
3779:PMID
3598:ISSN
3572:PMID
3521:PMID
3470:PMID
3452:ISSN
3405:PMID
3397:ISSN
3347:ISSN
3321:PMID
3313:ISSN
3207:PMID
3101:2021
2949:ISSN
2722:CASP
2693:PMID
2637:2020
2612:2020
2525:PMID
2507:ISSN
2425:See
2401:PMID
2393:ISSN
2356:2020
2331:2020
2253:ISSN
2209:ISSN
2174:PMID
1995:ISSN
1966:PMID
1948:ISSN
1716:PMID
1613:PMID
1505:2022
1455:CNBC
1402:2020
1105:The
891:CASP
733:ions
731:and
671:edge
592:GPUs
582:, a
556:CASP
543:and
499:and
485:ions
4763:NMT
4646:OCR
4641:HWR
4593:JAX
4547:VPU
4542:TPU
4537:IPU
4361:SGD
3787:),
3769:PMC
3759:doi
3721:at
3712:on
3562:PMC
3552:doi
3511:PMC
3501:doi
3460:PMC
3444:doi
3387:doi
3305:doi
3283:131
3197:doi
3185:608
2683:doi
2679:613
2515:PMC
2499:doi
2456:doi
2383:doi
2261:PMC
2243:doi
2164:doi
2152:604
1956:PMC
1940:doi
1928:596
1884:doi
1880:577
1857:doi
1706:PMC
1696:doi
1684:596
1605:doi
1593:588
1484:doi
1289:doi
1285:577
1262:doi
1074:yes
1070:Web
1059:yes
956:).
893:'s
803:on
721:RNA
717:DNA
618:how
560:GDT
477:RNA
473:DNA
193:Art
5183::
3777:.
3767:.
3755:19
3753:.
3749:.
3652:.
3627:.
3616:^
3596:.
3592:.
3570:.
3560:.
3548:28
3546:.
3542:.
3519:.
3509:.
3497:20
3495:.
3491:.
3468:.
3458:.
3450:.
3440:13
3434:.
3411:.
3403:.
3395:.
3383:28
3381:.
3377:.
3345:.
3341:.
3319:.
3311:.
3303:.
3295:.
3281:.
3277:.
3238:.
3227:^
3213:.
3205:.
3195:.
3183:.
3179:.
3160:.
3136:.
3117:.
3092:.
3068:.
3025:,
3002:.
2976:.
2947:.
2943:.
2918:.
2900:,
2883:,
2879:,
2861:.
2842:,
2823:,
2819:,
2807:).
2791:,
2772:,
2754:,
2737:,
2699:.
2691:.
2677:.
2673:.
2650:Ă….
2628:.
2602:.
2589:^
2573:.
2548:.
2537:^
2523:.
2513:.
2505:.
2497:.
2485:87
2483:.
2479:.
2452:87
2450:,
2446:,
2407:.
2399:.
2391:.
2379:17
2377:.
2373:.
2347:.
2322:.
2310:^
2259:.
2251:.
2237:.
2207:.
2203:.
2180:.
2172:.
2162:.
2150:.
2146:.
2097:,
2035:^
2024:,
2013:^
1989:.
1978:^
1964:.
1954:.
1946:.
1938:.
1926:.
1922:.
1910:^
1901:,
1873:,
1853:35
1851:,
1847:,
1819:.
1806:^
1789:.
1776:^
1760:.
1736:.
1714:.
1704:.
1694:.
1682:.
1678:.
1665:^
1655:.
1619:.
1611:.
1603:.
1591:.
1571:^
1547:^
1536:,
1513:^
1492:.
1482:.
1478:.
1429:.
1410:^
1393:.
1278:,
1258:87
1253:,
928:,
922:,
735:.
723:,
719:,
594:.
588:AI
539:,
487:.
475:,
464:.
4276:e
4269:t
4262:v
4190:"
4186:"
3818:e
3811:t
3804:v
3785:.
3761::
3743:(
3662:.
3638:.
3610:.
3578:.
3554::
3527:.
3503::
3476:.
3446::
3419:.
3389::
3359:.
3327:.
3307::
3299::
3289::
3242:.
3221:.
3199::
3191::
3146:.
3121:.
3103:.
3078:.
3012:.
2987:.
2961:.
2928:.
2707:.
2685::
2639:.
2614:.
2583:.
2559:.
2531:.
2501::
2458::
2442:,
2415:.
2385::
2358:.
2333:.
2269:.
2245::
2221:.
2188:.
2166::
2158::
2086:-
2007:.
1972:.
1942::
1934::
1886::
1859::
1830:.
1800:.
1770:.
1746:.
1722:.
1698::
1690::
1659:.
1627:.
1607::
1599::
1507:.
1486::
1463:.
1439:.
1404:.
1291::
1264::
1158:.
835:Ă…
765:Ă…
388:e
381:t
374:v
284:/
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
_artificial_intelligence_icon.png){kind=icon}