How to predict structures with AlphaFold - Revision history

Eric Martz: /* Single chain limitation */

Eric Martz — Sat, 09 Dec 2023 22:36:56 GMT

Single chain limitation

←Older revision		Revision as of 22:36, 9 December 2023
Line 12:		Line 12:

	==Single chain limitation==		==Single chain limitation==
-	Initially, AlphaFold and ColabFold performed best with single chains<ref name="complexes">[https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1 Protein complex prediction with AlphaFold-Multimer], 2021, Evans ''et al.'' (DeepMind Team).</ref>, which may include one or a few domains. The instructions below were written before ~~the AlphaFold Colab~~ was adapted to prediction of complexes or alternate conformations.	+	Initially, AlphaFold and ColabFold performed best with single chains<ref name="complexes">[https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1 Protein complex prediction with AlphaFold-Multimer], 2021, Evans ''et al.'' (DeepMind Team).</ref>, which may include one or a few domains. The instructions below were written before ColabFold was adapted to prediction of complexes or alternate conformations.
	If you are interested in complexes or alternate conformations, please see ColabFold instructions in the 2023 paper by Kim ''et al.'' <ref name="kim2023">[https://protocolexchange.researchsquare.com/article/pex-2490/v1 Easy and accurate protein structure prediction using ColabFold], 2023, Kim ''et al.'' (DeepMind Team).</ref>		If you are interested in complexes or alternate conformations, please see ColabFold instructions in the 2023 paper by Kim ''et al.'' <ref name="kim2023">[https://protocolexchange.researchsquare.com/article/pex-2490/v1 Easy and accurate protein structure prediction using ColabFold], 2023, Kim ''et al.'' (DeepMind Team).</ref>

Eric Martz: /* Single chain limitation */

Eric Martz — Sat, 09 Dec 2023 22:36:16 GMT

Single chain limitation

←Older revision		Revision as of 22:36, 9 December 2023
Line 12:		Line 12:

	==Single chain limitation==		==Single chain limitation==
-	AlphaFold ~~performs~~ best with single chains~~, which may include one or a few domains. It is less successful predicting homo- or hetero-complexes~~<ref name="complexes">[https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1 Protein complex prediction with AlphaFold-Multimer], 2021, Evans ''et al.'' (DeepMind Team).</ref>. ~~Such attempts require special maneuvers that are beyond~~ the ~~scope~~ of ~~the following~~ instructions.	+	Initially, AlphaFold and ColabFold performed best with single chains<ref name="complexes">[https://www.biorxiv.org/content/10.1101/2021.10.04.463034v1 Protein complex prediction with AlphaFold-Multimer], 2021, Evans ''et al.'' (DeepMind Team).</ref>, which may include one or a few domains. The instructions below were written before the AlphaFold Colab was adapted to prediction of complexes or alternate conformations.
		+	If you are interested in complexes or alternate conformations, please see ColabFold instructions in the 2023 paper by Kim ''et al.'' <ref name="kim2023">[https://protocolexchange.researchsquare.com/article/pex-2490/v1 Easy and accurate protein structure prediction using ColabFold], 2023, Kim ''et al.'' (DeepMind Team).</ref>

	==Submitting A Sequence==		==Submitting A Sequence==

Eric Martz at 00:11, 29 September 2023

Eric Martz — Fri, 29 Sep 2023 00:11:28 GMT

←Older revision		Revision as of 00:11, 29 September 2023
Line 27:		Line 27:
	[[Image:AF2Adv-seq1.png\|400px]]		[[Image:AF2Adv-seq1.png\|400px]]
	<br>		<br>
-	This input slot can accept sequences >1,000 amino acids, even though it is only one line. Sequence lengths of ~1,000 amino acids, or longer, may cause the Colab to fail, but can be predicted by submitting in two halves.<ref name="halves">I had one sequence of length ~1,300. After it failed, I submitted it as two halves with a substantial overlap (~350 residues). The middle overlap of ~200 residues of the predicted structures superposed very closely with [[DeepView]]. I trimmed off the ends that superposed poorly, and superposed the two halves via the mid-overlap. By inspection, I chose pair of alpha carbons near the middle where the alpha carbon positions were nearly identical. I trimmed each half to this position, and "ligated" the two halves by combining the superposed half PDB files with a text editor. For further details, contact [[User:Eric_Martz]].</ref>	+	This input slot can accept sequences >1,000 amino acids, even though it is only one line. Sequence lengths of ~1,000 amino acids, or longer, may cause the Colab to fail, but can be predicted by submitting in two halves.<ref name="halves">I had one sequence of length ~1,300. After it failed, I submitted it as two halves with a substantial overlap (~350 residues). The middle overlap of ~200 residues of the predicted structures superposed very closely with [[DeepView]]. I trimmed off the ends that superposed poorly, and superposed the two halves via the mid-overlap. By inspection, I chose pair of alpha carbons near the middle where the alpha carbon positions were nearly identical. I trimmed each half to this position, and "ligated" the two halves by combining the superposed half PDB files with a text editor. For further details, contact [[User:Eric_Martz]].</ref> See also <ref name="halves" /> and [[Joining AlphaFold predictions for halves of a molecule]].
	<br><br>		<br><br>
	4. Enter a jobname in the slot below the sequence slot. The results.zip filename will begin with this jobname (but none of its contents include the jobname).		4. Enter a jobname in the slot below the sequence slot. The results.zip filename will begin with this jobname (but none of its contents include the jobname).

Eric Martz: /* References and Notes */

Eric Martz — Fri, 29 Sep 2023 00:05:51 GMT

References and Notes

←Older revision		Revision as of 00:05, 29 September 2023
Line 100:		Line 100:

	In October, 2021, a subscription to [https://colab.research.google.com/signup/pricing Colab Pro] was US $10/month. After I subscribed, all jobs I have submitted have been processed without restriction.		In October, 2021, a subscription to [https://colab.research.google.com/signup/pricing Colab Pro] was US $10/month. After I subscribed, all jobs I have submitted have been processed without restriction.
		+
		+	==See Also==
		+	*[[AlphaFold/Index]], a list of pages in Proteopedia about Alphafold.

	==References and Notes==		==References and Notes==
	<references />		<references />

Eric Martz at 23:41, 16 August 2023

Eric Martz — Wed, 16 Aug 2023 23:41:31 GMT

←Older revision		Revision as of 23:41, 16 August 2023
Line 1:		Line 1:
-	In 2020, the [[AlphaFold]] project of Google's DeepMind team demonstrated a major breakthrough in predicting protein structure from sequence. Their success in the blind [[CASP]] competition astonished many experts. For an overview, see [[Theoretical models]], bearing in mind "The Joys and Perils of AlphaFold"<ref name="perils">PMID: 34668287</ref>.	+	In 2020, the [[AlphaFold]] project of Google's DeepMind team demonstrated a major breakthrough in predicting protein structure from sequence. Their success in the blind [[CASP]] competition astonished many experts. For an overview, see [[Theoretical models]], bearing in mind "The Joys and Perils of AlphaFold"<ref name="perils">PMID: 34668287</ref>. AlphaFold2 continued to have the highest success rate in the [[Theoretical_models#2022:_CASP_15\|2022 CASP 15 competition]].

	In July, 2021, DeepMind released [[Alphafold#AlphaFold_published_July_2021\|AlphaFold as open source code]]. Subsequently, [[Alphafold#Free_AlphaFold-based_Servers\|several Colabs became available]] offering '''free''' structure prediction for user-submitted protein sequences. These Google Colabs (collaboratories)<ref name="colabfaq">[https://research.google.com/colaboratory/faq.html Collaboratory FAQ] at Google.</ref>. enable users to submit sequences via web browser, executing the code in the Google cloud, using space private to each user, returning predicted structures.		In July, 2021, DeepMind released [[Alphafold#AlphaFold_published_July_2021\|AlphaFold as open source code]]. Subsequently, [[Alphafold#Free_AlphaFold-based_Servers\|several Colabs became available]] offering '''free''' structure prediction for user-submitted protein sequences. These Google Colabs (collaboratories)<ref name="colabfaq">[https://research.google.com/colaboratory/faq.html Collaboratory FAQ] at Google.</ref>. enable users to submit sequences via web browser, executing the code in the Google cloud, using space private to each user, returning predicted structures.

Eric Martz at 23:38, 16 August 2023

Eric Martz — Wed, 16 Aug 2023 23:38:42 GMT

←Older revision		Revision as of 23:38, 16 August 2023
Line 1:		Line 1:
-	In 2020, the AlphaFold project of Google's DeepMind team demonstrated a major breakthrough in predicting protein structure from sequence. Their success in the blind [[CASP]] competition astonished many experts. For an overview, see [[Theoretical models]], bearing in mind "The Joys and Perils of AlphaFold"<ref name="perils">PMID: 34668287</ref>.	+	In 2020, the [[AlphaFold]] project of Google's DeepMind team demonstrated a major breakthrough in predicting protein structure from sequence. Their success in the blind [[CASP]] competition astonished many experts. For an overview, see [[Theoretical models]], bearing in mind "The Joys and Perils of AlphaFold"<ref name="perils">PMID: 34668287</ref>.

	In July, 2021, DeepMind released [[Alphafold#AlphaFold_published_July_2021\|AlphaFold as open source code]]. Subsequently, [[Alphafold#Free_AlphaFold-based_Servers\|several Colabs became available]] offering '''free''' structure prediction for user-submitted protein sequences. These Google Colabs (collaboratories)<ref name="colabfaq">[https://research.google.com/colaboratory/faq.html Collaboratory FAQ] at Google.</ref>. enable users to submit sequences via web browser, executing the code in the Google cloud, using space private to each user, returning predicted structures.		In July, 2021, DeepMind released [[Alphafold#AlphaFold_published_July_2021\|AlphaFold as open source code]]. Subsequently, [[Alphafold#Free_AlphaFold-based_Servers\|several Colabs became available]] offering '''free''' structure prediction for user-submitted protein sequences. These Google Colabs (collaboratories)<ref name="colabfaq">[https://research.google.com/colaboratory/faq.html Collaboratory FAQ] at Google.</ref>. enable users to submit sequences via web browser, executing the code in the Google cloud, using space private to each user, returning predicted structures.

Eric Martz: /* First Check AlphaFold Database */

Eric Martz — Mon, 29 Nov 2021 20:37:33 GMT

First Check AlphaFold Database

←Older revision		Revision as of 20:37, 29 November 2021
Line 8:		Line 8:
	==First Check AlphaFold Database==		==First Check AlphaFold Database==
	Structure predictions for over 300,000 proteins are already available in the [https://alphafold.ebi.ac.uk/ AlphaFold Database]. If your protein is there, you don't need to proceed with the instructions below. Simply download the prediction from the Database.		Structure predictions for over 300,000 proteins are already available in the [https://alphafold.ebi.ac.uk/ AlphaFold Database]. If your protein is there, you don't need to proceed with the instructions below. Simply download the prediction from the Database.
		+
		+	Also check the [http://alphafill.eu AlphaFill Database], which has added ligands to appropriate AlphaFold predictions. Ligand positioning is approximate. See [[Alphafold#Ligands:_AlphaFill \|CAUTION]] provided by the AlphaFill team.

	==Single chain limitation==		==Single chain limitation==

Eric Martz: /* Single chain limitation */

Eric Martz — Mon, 29 Nov 2021 20:33:50 GMT

Single chain limitation

←Older revision		Revision as of 20:33, 29 November 2021
Line 5:		Line 5:
	Below are instructions for beginners who wish to predict structures. We recommend the "advanced" Colab by [https://www.biorxiv.org/content/10.1101/2021.08.15.456425v1 Sergey Ovchinnikov, Milot Mirdita and Martin Steinegger]. Some of the options recommended below were gleaned from the [https://www.youtube.com/watch?v=Rfw7thgGTwI 1 hour 46 minute video of presentations by Sergey Ovchinnikov and Martin Steinegger] (August, 2021) for the Boston Protein Design and Modeling Club		Below are instructions for beginners who wish to predict structures. We recommend the "advanced" Colab by [https://www.biorxiv.org/content/10.1101/2021.08.15.456425v1 Sergey Ovchinnikov, Milot Mirdita and Martin Steinegger]. Some of the options recommended below were gleaned from the [https://www.youtube.com/watch?v=Rfw7thgGTwI 1 hour 46 minute video of presentations by Sergey Ovchinnikov and Martin Steinegger] (August, 2021) for the Boston Protein Design and Modeling Club
	hosted by Chris Bahl.		hosted by Chris Bahl.
		+
		+	==First Check AlphaFold Database==
		+	Structure predictions for over 300,000 proteins are already available in the [https://alphafold.ebi.ac.uk/ AlphaFold Database]. If your protein is there, you don't need to proceed with the instructions below. Simply download the prediction from the Database.

	==Single chain limitation==		==Single chain limitation==

Eric Martz: /* Estimated Reliability */

Eric Martz — Mon, 22 Nov 2021 17:54:10 GMT

Estimated Reliability

←Older revision		Revision as of 17:54, 22 November 2021
Line 50:		Line 50:
	===Estimated Reliability===		===Estimated Reliability===

-	Each predicted model has an average estimated reliability (pLDDT). >90 is likely accurate; <70 is low confidence. For more about interpreting these values, please see the [https://alphafold.ebi.ac.uk/faq AlphaFold Database FAQ].	+	Each predicted model has an average estimated reliability (pLDDT, predicted local distance difference test). >90 is likely accurate; <70 is low confidence. For more about interpreting these values, please see the [https://alphafold.ebi.ac.uk/faq AlphaFold Database FAQ].

	Each residue has an estimated reliability of its position (0-100) in the PDB [[temperature]] column. BEWARE that high values mean high confidence, and low values mean low confidence. This is the INVERSE of [[temperature\|crystallographic temperature values]], where low values are good and high values are bad. Uploading your PDB file to [http://firstglance.jmol.org FirstGlance in Jmol] will automatically color each residue by its estimated reliability.		Each residue has an estimated reliability of its position (0-100) in the PDB [[temperature]] column. BEWARE that high values mean high confidence, and low values mean low confidence. This is the INVERSE of [[temperature\|crystallographic temperature values]], where low values are good and high values are bad. Uploading your PDB file to [http://firstglance.jmol.org FirstGlance in Jmol] will automatically color each residue by its estimated reliability.

Eric Martz: /* Estimated Reliability */

Eric Martz — Sat, 23 Oct 2021 20:52:49 GMT

Estimated Reliability

←Older revision		Revision as of 20:52, 23 October 2021
Line 52:		Line 52:
	Each predicted model has an average estimated reliability (pLDDT). >90 is likely accurate; <70 is low confidence. For more about interpreting these values, please see the [https://alphafold.ebi.ac.uk/faq AlphaFold Database FAQ].		Each predicted model has an average estimated reliability (pLDDT). >90 is likely accurate; <70 is low confidence. For more about interpreting these values, please see the [https://alphafold.ebi.ac.uk/faq AlphaFold Database FAQ].

-	Each residue has an estimated reliability of its position (0-100) in the PDB [[temperature]] column. BEWARE that high values mean high confidence, and low values mean low confidence. This is the INVERSE of [[temperature\|crystallographic temperature values]], where low values are good and high values are bad.	+	Each residue has an estimated reliability of its position (0-100) in the PDB [[temperature]] column. BEWARE that high values mean high confidence, and low values mean low confidence. This is the INVERSE of [[temperature\|crystallographic temperature values]], where low values are good and high values are bad. Uploading your PDB file to [http://firstglance.jmol.org FirstGlance in Jmol] will automatically color each residue by its estimated reliability.

	===Visualization===		===Visualization===