AlphaFold - Revision history

Eric Martz at 14:52, 29 September 2026

2026-09-29T14:52:26Z

← Older revision		Revision as of 14:52, 29 September 2026
Line 19:		Line 19:
	* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.		* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.
	* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].		* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].
	* Dropping the predicted model into [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [https://molviz.org/firstglance/fgij/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].		* Dropping the predicted model into [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [https://molviz.org/firstglance/fgij/fg.htm?mol=AFDB-Q9AY27-F1-model_v1.pdb Interactive].
	</td></tr></table>		</td></tr></table>

Eric Martz at 14:48, 29 September 2026

2026-09-29T14:48:25Z

← Older revision		Revision as of 14:48, 29 September 2026
Line 19:		Line 19:
	* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.		* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.
	* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].		* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].
	* Dropping the predicted model into [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [~~http~~://~~firstglance.jmol~~.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].		* Dropping the predicted model into [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [https://molviz.org/firstglance/fgij/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].
	</td></tr></table>		</td></tr></table>

Eric Martz at 14:46, 29 September 2026

2026-09-29T14:46:52Z

← Older revision		Revision as of 14:46, 29 September 2026
Line 19:		Line 19:
	* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.		* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.
	* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].		* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].
	* ~~Uploading~~ the predicted ~~PDB file to~~ [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [http://firstglance.jmol.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].		* Dropping the predicted model into [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [http://firstglance.jmol.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].
	</td></tr></table>		</td></tr></table>

Eric Martz at 17:08, 10 November 2024

2024-11-10T17:08:02Z

← Older revision		Revision as of 17:08, 10 November 2024
Line 22:		Line 22:
	</td></tr></table>		</td></tr></table>

			2021 Resources: [For more recent resources and servers, see [[How to predict structures with AlphaFold]].]
	* See [https://www.ebi.ac.uk/training/online/courses/alphafold AlphaFold A practical Guide] Superb EMBL-EBI Interactive online tutorial on AlphaFold2 (~3 hours)		* See [https://www.ebi.ac.uk/training/online/courses/alphafold AlphaFold A practical Guide] Superb EMBL-EBI Interactive online tutorial on AlphaFold2 (~3 hours)
	*See short [https://mediasite.embl.de/Mediasite/Play/a320afff218d4a3cbad6ea6eca5212931d superb lecture] on AlphaFold by the CEO of ''DeepMind'', '''Dennis Hassabis''', that was given at the EMBL, Heidelberg, on 3-Feb-2022, entitled '''Using AI to accelerate scientific discovery'''.		*See short [https://mediasite.embl.de/Mediasite/Play/a320afff218d4a3cbad6ea6eca5212931d superb lecture] on AlphaFold by the CEO of ''DeepMind'', '''Dennis Hassabis''', that was given at the EMBL, Heidelberg, on 3-Feb-2022, entitled '''Using AI to accelerate scientific discovery'''.

Eric Martz: /* AlphaFold published July 2021 */

2024-11-10T17:06:46Z

AlphaFold published July 2021

← Older revision		Revision as of 17:06, 10 November 2024
Line 54:		Line 54:

	==AlphaFold published July 2021==		==AlphaFold published July 2021==
			[For recent prediction servers, see [[How to predict structures with AlphaFold]].]

	AlphaFold was published in July, 2021<ref name="af2021">PMID: 34265844</ref>. Methods were described in considerable detail. The source code, trained weights, and inference script were made available under an '''open-source license'''. Structure prediction required about one GPU (Graphics Processing Unit) minute per model of about 384 amino acids.		AlphaFold was published in July, 2021<ref name="af2021">PMID: 34265844</ref>. Methods were described in considerable detail. The source code, trained weights, and inference script were made available under an '''open-source license'''. Structure prediction required about one GPU (Graphics Processing Unit) minute per model of about 384 amino acids.

Eric Martz: /* See Also */

2024-11-10T16:55:31Z

Eric Martz: /* Advances since 2021 */

2024-11-10T16:52:53Z

Advances since 2021

← Older revision		Revision as of 16:52, 10 November 2024
Line 101:		Line 101:

	==Advances since 2021==		==Advances since 2021==
			*See a list of newer servers at [[How to predict structures with AlphaFold]].
	*RoseTTAFoldNA<ref>PMID: 37996753</ref> offers a leap forward in predicting structures of complexes of proteins and nucleic acids, but in November 2023 is not yet available as a free server.		*RoseTTAFoldNA<ref>PMID: 37996753</ref> offers a leap forward in predicting structures of complexes of proteins and nucleic acids, but in November 2023 is not yet available as a free server.

Eric Martz: /* Free AlphaFold-based Servers */

2024-11-10T16:50:50Z

Free AlphaFold-based Servers

← Older revision		Revision as of 16:50, 10 November 2024
Line 67:		Line 67:
	<table style="background-color:#e0ffe0;border:1px solid black;font-size:120%;"><tr><td>		<table style="background-color:#e0ffe0;border:1px solid black;font-size:120%;"><tr><td>
	If you want an AlphaFold-predicted structure for a protein sequence:		If you want an AlphaFold-predicted structure for a protein sequence:
	* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it.		* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.
	* Otherwise, follow [[How to predict structures with AlphaFold]].		* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].
	* Uploading the predicted PDB file to [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png ~~Example~~].		* Uploading the predicted PDB file to [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [http://firstglance.jmol.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].
	</td></tr></table>		</td></tr></table>

Eric Martz at 16:45, 10 November 2024

2024-11-10T16:45:02Z

← Older revision		Revision as of 16:45, 10 November 2024
Line 17:		Line 17:
	<table style="background-color:#e0ffe0;border:1px solid black;font-size:120%;"><tr><td>		<table style="background-color:#e0ffe0;border:1px solid black;font-size:120%;"><tr><td>
	If you want an AlphaFold-predicted structure for a protein sequence:		If you want an AlphaFold-predicted structure for a protein sequence:
	* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it.		* If a prediction is already in the [[#AlphaFold Database of Predictions\|AlphaFold Database]], simply download it. Limited to single chain proteins without ligands.
	* Otherwise, follow [[How to predict structures with AlphaFold]].		* Otherwise, and for multiple chain (protein/nucleic acid) structures with ligands, follow [[How to predict structures with AlphaFold]].
	* Uploading the predicted PDB file to [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [http://firstglance.jmol.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].		* Uploading the predicted PDB file to [http://FirstGlance.Jmol.Org FirstGlance in Jmol] will automatically '''color it by estimated reliability per residue'''. Examples: [http://bioinformatics.org/molvis/images/firstglance-with-alphafold.png Snapshot], [http://firstglance.jmol.org/fg.htm?mol=AF-Q9AY27-F1-model_v1.pdb Interactive].
	</td></tr></table>		</td></tr></table>

Eric Martz at 14:35, 10 November 2024

2024-11-10T14:35:49Z

← Older revision		Revision as of 14:35, 10 November 2024
Line 1:		Line 1:
	[[Image:Hassabis_Demis.jpg\|300px\|right\|thumb\| Demis Hassabis - DeepMind]]		[[Image:Hassabis_Demis.jpg\|300px\|right\|thumb\| Demis Hassabis - DeepMind]]
	[[Image:Jumper_John.jpg\|300px\|right\|thumb\| John Jumper - DeepMind]]		[[Image:Jumper_John.jpg\|300px\|right\|thumb\| John Jumper - DeepMind]]
	<table style="background-color:#~~c000c0~~;"><tr><td>		<table style="background-color:#ffffb0;"><tr><td>
	October 2024: David Baker, Demis Hassabis, and John M. Jumper share the [[Nobel_Prizes_for_3D_Molecular_Structure#2020-2029\|Nobel Prize in Chemistry]], Baker for "computational protein design", Hassabis and Jumper for "protein structure prediction", namely, AlphaFold.		October 2024: David Baker, Demis Hassabis, and John M. Jumper share the [[Nobel_Prizes_for_3D_Molecular_Structure#2020-2029\|Nobel Prize in Chemistry]], Baker for "computational protein design", Hassabis and Jumper for "protein structure prediction", namely, AlphaFold.
	* [https://www.youtube.com/watch?v=cx7l9ZGFZkw 22 min video] explaining their contributions.		* [https://www.youtube.com/watch?v=cx7l9ZGFZkw 22 min video] explaining their contributions.

← Older revision		Revision as of 16:55, 10 November 2024
Line 105:		Line 105:

	==See Also==		==See Also==
			*[[How to predict structures with AlphaFold]].
	*[[AlphaFold/Index]], a list of pages in Proteopedia about Alphafold.		*[[AlphaFold/Index]], a list of pages in Proteopedia about Alphafold.
			*[[How To Find A Structure]] covers both [[empirical models]] and the advantages of comparing them with AlphaFold models due to [[missing residues and incomplete sidechains]] prevalent in empirical models.

	==References==		==References==