Flexible proteins

Introduction

AlphaFold 3 is a major advance because biology is fundamentally about interaction. It can often predict how proteins, DNA, RNA and small molecules fit together far better than earlier AI systems. But one of the biggest remaining problems is that many proteins are not fixed objects at all. They bend, fluctuate, partially unfold, switch states and form temporary interactions that exist only for fractions of a second. [Nature]nature.comNatureAccurate structure prediction of biomolecular interactions…by J Abramson · 2024 · Cited by 14421 — Here we describe our AlphaFol… [ebi.ac.uk]ebi.ac.ukWhat AlphaFold 3 struggles withNevertheless, AlphaFold 3 still has limitations. A key limitation of protein structure prediction models i…

Overview image for Flexible proteins That matters because a static structure can look highly convincing while still missing the biology that determines whether a drug actually works inside a living cell. In many cases, proteins behave less like rigid locks and more like moving ensembles of shapes. A single predicted structure may capture one frame of the film while missing the rest of the movie. [LinkedIn]linkedin.comLinkedInAlphaFold limitations in drug developmentEven with AlphaFold, protein structure is not a solved problem. And protein structure wa… [OUP Academic]academic.oup.comOUP AcademicBeyond static structures: protein dynamic conformations…by X Cui · 2025 · Cited by 37 — This review outlines the fundament…

This flexibility problem is one reason AI-driven biology remains both exciting and incomplete. The same systems that may dramatically accelerate medicine and scientific discovery still struggle with some of the deepest realities of molecular life: motion, uncertainty and context.

Why proteins behave more like movies than snapshots

For decades, structural biology often focused on finding one “correct” protein structure. X-ray crystallography, cryo-electron microscopy and related techniques usually aim to capture stable conformations. AI systems trained heavily on those datasets naturally inherit that bias toward fixed structures. [ebi.ac.uk]ebi.ac.ukstrengths and limitations of alphafold25 Jan 2024 — AlphaFold2 can be used to identify intrinsically disordered regions. Naturally, the system cannot predict disordered or dyn… [Nature]nature.comHowever, most molecular docking…Read more…

But proteins inside cells are constantly moving.

Some proteins open and close like mechanical jaws. Others fluctuate between active and inactive states. Ion channels embedded in cell membranes switch conformations in milliseconds. Enzymes may briefly adopt rare shapes that become crucial during catalysis or drug binding. Even apparently stable proteins often “breathe”, with loops and side chains shifting continuously. [Nature]nature.comNatureAlphaFold predictions are valuable hypotheses and…by TC Terwilliger · 2024 · Cited by 391 — We evaluate how well AlphaFold predi… [MDPI]mdpi.comMolecular Dynamics Simulations in Drug Discovery and…by OMH Salo-Ahen · 2020 · Cited by 677 — Protein conformation is one of the bigge…

In modern biophysics, many researchers now describe proteins as conformational ensembles rather than single structures. The biologically relevant object is not one shape but a probability distribution across many possible states. [OUP Academic]academic.oup.comOUP AcademicBeyond static structures: protein dynamic conformations…by X Cui · 2025 · Cited by 37 — This review outlines the fundament… 2arXiv

That creates a difficult mismatch with static prediction systems.

AlphaFold 3 can often generate highly accurate molecular arrangements, but EMBL-EBI’s own training material notes that the model still predicts “a single structure” rather than the full dynamic behaviour of molecules in solution. [ebi.ac.uk]ebi.ac.ukHow have AlphaFold 3's predictions been validated?This study found that AF3 excels at predicting static protein-ligand interactions where…

This distinction sounds technical, but it changes how scientists interpret AI predictions:

A predicted binding pocket may only exist briefly in reality.
A drug may bind strongly to one conformation but weakly to another.
A protein may switch states after binding.
Cellular conditions such as pH, crowding, phosphorylation or membrane environment may reshape the interaction entirely.

In other words, biology is often governed by transitions between states rather than by any one frozen configuration.

Flexible proteins illustration 1

Disordered regions, allostery and transient binding

Intrinsically disordered proteins challenge the whole idea of a single structure

One of the strongest examples of the flexibility problem comes from intrinsically disordered proteins and intrinsically disordered regions, often abbreviated as IDPs and IDRs.

These proteins or segments do not settle into one stable folded structure under normal conditions. Instead, they remain highly flexible and may adopt different conformations depending on binding partners or cellular context. They are especially common in signalling, transcription regulation and other control systems central to complex organisms. [ebi.ac.uk]ebi.ac.ukWhat AlphaFold 3 struggles withNevertheless, AlphaFold 3 still has limitations. A key limitation of protein structure prediction models i… [ScienceDirect]sciencedirect.comAlphaFold and Implications for Intrinsically Disordered…by KM Ruff · 2021 · Cited by 714 — AlphaFold, a deep learning-based approach t…

This is not a minor edge case. Large fractions of the human proteome contain significant disordered regions. [arXiv]arxiv.orgSource details in endnotes.

That creates a serious challenge for structure-prediction AI. AlphaFold systems can often identify where disorder exists through low-confidence scores, but they cannot fully predict a single “correct” shape for proteins that may not possess one in the first place. EMBL-EBI explicitly notes that AlphaFold cannot predict structures “that do not exist in one set conformation in nature”. [ebi.ac.uk]ebi.ac.ukstrengths and limitations of alphafold25 Jan 2024 — AlphaFold2 can be used to identify intrinsically disordered regions. Naturally, the system cannot predict disordered or dyn…

Researchers are now trying to move beyond this limitation using ensemble approaches that combine AI predictions with molecular dynamics simulations and experimental measurements such as nuclear magnetic resonance spectroscopy. Nature [PMC]pmc.ncbi.nlm.nih.govPMCAccurate Protein Dynamic Conformational EnsemblesProtein Dynamic Conformational Ensembles - PMCby D Lesovoy · 2025 · Cited by 2 — In this study, we present an approach that integrates fr…

The need for these hybrid methods reveals an important point about AI-enabled scientific acceleration. AI can compress huge amounts of structural inference, but living systems may still require probabilistic, dynamic and experimentally grounded models rather than clean deterministic answers.

Allostery means distant parts of proteins can control each other

Another major challenge for static models is allostery.

Allostery occurs when a molecule binds at one site on a protein but changes behaviour somewhere else, often far away. A small structural shift can ripple through the protein and alter activity at a distant functional region. [ScienceDirect]sciencedirect.comAlphaFold and Implications for Intrinsically Disordered…by KM Ruff · 2021 · Cited by 714 — AlphaFold, a deep learning-based approach t… [2eLife]elifesciences.orgDrug Discovery: Decoding the mechanisms of allosteryby S Khan · 2023 · Cited by 3 — A complex interplay between structure, conformational…

This is central to modern pharmacology because many important drugs work allosterically rather than by simply blocking an active site.

The difficulty is that allosteric regulation is inherently dynamic. The crucial information may lie not in one structure but in how populations of structures shift over time. A protein might transiently expose hidden pockets that exist only in rare conformations. [Nature]nature.comHighly accurate protein structure prediction with AlphaFoldby J Jumper · 2021 · Cited by 49928 — The AlphaFold network directly predicts… [MDPI]mdpi.comEvaluating AlphaFold 3 Folding of the Intrinsically…by CM Nartey · 2025 — Background/Objectives: Intrinsically disordered protein regi…

Static prediction systems can therefore miss some of the most medically valuable biology.

A protein structure may appear complete while concealing the fact that the therapeutically important state is:

rare,
short-lived,
ligand-induced,
or dependent on cellular conditions absent from the model.

This is one reason drug discovery still relies heavily on molecular dynamics simulations, biophysical experiments and iterative laboratory validation even in the AlphaFold era. [Nature]nature.comNatureAlphaFold prediction of structural ensembles of disordered…by ZF Brotzakis · 2025 · Cited by 121 — We introduce the AlphaFold-Me… [MDPI]mdpi.comAI-Driven Design of Miniproteins as Potential Allosteric…by X Liu · 2026 — Allosteric sites are often shallow, transient, or only popu…

Some binding events only exist briefly

Many biologically important interactions are transient rather than stable.

Signalling proteins may touch only briefly before separating. Regulatory complexes may assemble temporarily under stress conditions. Flexible peptide regions can bind weakly and reversibly, switching between many poses. [PNAS]pnas.orgPNASAlphaFold-Multimer accurately captures interactions and…by A Omidi · 2024 · Cited by 75 — Our study demonstrates that AlphaFold-Mu…

These interactions are difficult because there may be no single dominant geometry to predict. Instead, the biologically relevant behaviour emerges statistically across many fleeting encounters.

AlphaFold-derived systems have shown promising results on some flexible peptide interactions and disordered binding regions. [PNAS]pnas.orgPNASAlphaFold-Multimer accurately captures interactions and…by A Omidi · 2024 · Cited by 75 — Our study demonstrates that AlphaFold-Mu… But even optimistic studies usually frame these outputs as approximations or hypotheses rather than final mechanistic truth.

That caution matters. In biology, a visually persuasive structure can create false confidence if readers forget that the underlying system is probabilistic and dynamic.

Flexible proteins illustration 2

How missing dynamics can mislead drug design

A good-looking binding pose may not survive reality

One of the biggest practical risks in AI-guided drug discovery is over-interpreting static docking predictions.

A model may place a small molecule neatly inside a protein pocket with apparently excellent geometric complementarity. But inside cells, both the ligand and the protein fluctuate continuously. Solvent molecules interfere. Side chains rearrange. Entire domains may shift. [Nature]nature.comStructure prediction of alternative protein conformationsby P Bryant · 2024 · Cited by 113 — Here, we train a structure prediction networ… [arXiv]arxiv.orgOpen source on arxiv.org.

This means a predicted interaction that appears chemically plausible in one frame may:

collapse under thermal motion,
fail because the pocket rarely exists,
or trigger conformational changes that invalidate the original pose.

Drug developers have faced versions of this problem long before AI. Traditional docking systems frequently treated proteins as rigid objects because modelling flexibility is computationally expensive. Reviews of structure-based drug discovery repeatedly identify target flexibility as one of the field’s hardest unresolved problems. [Nature]nature.comprotein binder design and conformational state predictionby LA Abriata · 2026 — The communities of structural biology, molecular modeling…

AlphaFold 3 improves structure prediction substantially, but it does not eliminate this deeper physical difficulty.

Ligand binding can reshape the protein itself

Another complication is induced fit.

Sometimes a protein changes shape because a ligand binds. The binding event itself reshapes the target. Static predictions generated without the ligand may therefore miss the biologically relevant state entirely. [ScienceDirect]sciencedirect.comAlphaFold and Implications for Intrinsically Disordered…by KM Ruff · 2021 · Cited by 714 — AlphaFold, a deep learning-based approach t…

This is especially important for:

kinase inhibitors,
G protein-coupled receptors, [elifesciences.org]elifesciences.orgDrug Discovery: Decoding the mechanisms of allosteryby S Khan · 2023 · Cited by 3 — A complex interplay between structure, conformational…
ion channels,
and allosteric drugs.

Several recent studies now combine AlphaFold outputs with molecular dynamics simulations or flexible docking systems specifically to recover these ligand-induced conformations. [eLife]elifesciences.orgDrug Discovery: Decoding the mechanisms of allosteryby S Khan · 2023 · Cited by 3 — A complex interplay between structure, conformational… [2eLife]elifesciences.orgDrug Discovery: Decoding the mechanisms of allosteryby S Khan · 2023 · Cited by 3 — A complex interplay between structure, conformational…

The pattern is revealing. AI structure prediction is increasingly becoming the starting point for dynamic modelling rather than the endpoint.

Flexible proteins illustration 3

False certainty can become a scientific bottleneck

The visual clarity of AI-generated structures is both a strength and a danger.

A static protein image looks authoritative. Non-specialists may assume that once the structure is predicted, the biological mechanism is largely solved. But researchers repeatedly caution that AlphaFold predictions are hypotheses requiring experimental interpretation. [Nature]nature.comProc. Natl. Acad. Sci. USA 121…Read more…

This matters for the broader AI bloom discussion because accelerated science depends not only on generating outputs quickly, but on understanding uncertainty correctly.

The danger is not merely technical failure. It is epistemic overconfidence:

treating probabilistic states as fixed truths,
treating incomplete models as comprehensive explanations,
or underestimating how much biology still depends on context invisible to current AI systems.

If future AI systems become dramatically more capable, handling uncertainty in complex scientific systems may become as important as raw predictive power itself.

Why this limitation still fits the larger story of AI-enabled science

The flexibility problem does not negate the significance of AlphaFold 3. In many areas, the system already reduces enormous amounts of experimental search and may help accelerate medicine, protein engineering and molecular biology. [Nature]nature.comA comprehensive application of FiveFold for conformation…by SK Niazi · 2025 · Cited by 2 — We discuss future applications in structure…

But flexible proteins reveal something important about the frontier.

Biology is not simply a catalogue of static structures waiting to be decoded. It is a dynamic physical process unfolding across time, temperature, chemistry and cellular context. Predicting that process may require AI systems that model ensembles, energy landscapes and molecular trajectories rather than isolated snapshots. [OUP Academic]academic.oup.comOUP AcademicBeyond static structures: protein dynamic conformations…by X Cui · 2025 · Cited by 37 — This review outlines the fundament… [ScienceDirect]sciencedirect.comScienceDirectUnraveling structural mechanisms of allosteric drug actionby R Nussinov · 2014 · Cited by 141 — Our structural analysis reve…

That challenge also helps explain why many scientists view current AI biology as the beginning of a transition rather than the end of one. The field is moving from:

static prediction,
toward dynamic simulation,
probabilistic modelling,
and eventually integrated virtual-cell systems.

If those systems become reliable, the implications for the long-term future could be profound. Faster understanding of disease, ageing, signalling and molecular control systems could contribute to the broader possibility of AI-enabled scientific abundance and radical medical acceleration.

But flexible proteins are a reminder that living systems still contain layers of complexity that resist clean compression into a single elegant output. Even in an age of rapidly advancing AI, biology remains stubbornly alive, moving and context-dependent.

Endnotes

Source: nature.com
Link: https://www.nature.com/articles/s41586-024-07487-w
Source snippet
NatureAccurate structure prediction of biomolecular interactions...by J Abramson · 2024 · Cited by 14421 — Here we describe our AlphaFol...
Source: ebi.ac.uk
Link: https://www.ebi.ac.uk/training/online/courses/alphafold/alphafold-3-and-alphafold-server/introducing-alphafold-3/what-alphafold-3-struggles-with/
Source snippet
What AlphaFold 3 struggles withNevertheless, AlphaFold 3 still has limitations. A key limitation of protein structure prediction models i...
Source: nature.com
Link: https://www.nature.com/articles/s44386-024-00001-2
Source snippet
However, most molecular docking...Read more...
Source: linkedin.com
Link: https://www.linkedin.com/posts/%F0%9F%8E%AF-ming-tommy-tang-40650014_alphafolddrugdevelopment-activity-7436051840984576000-CzYG
Source snippet
LinkedInAlphaFold limitations in drug developmentEven with AlphaFold, protein structure is not a solved problem. And protein structure wa...
Source: academic.oup.com
Link: https://academic.oup.com/bib/article/26/4/bbaf340/8202937
Source snippet
OUP AcademicBeyond static structures: protein dynamic conformations...by X Cui · 2025 · Cited by 37 — This review outlines the fundament...
Source: mdpi.com
Link: https://www.mdpi.com/2227-9717/9/1/71
Source snippet
Molecular Dynamics Simulations in Drug Discovery and...by OMH Salo-Ahen · 2020 · Cited by 677 — Protein conformation is one of the bigge...
Source: nature.com
Link: https://www.nature.com/articles/s41592-023-02087-4
Source snippet
NatureAlphaFold predictions are valuable hypotheses and...by TC Terwilliger · 2024 · Cited by 391 — We evaluate how well AlphaFold predi...
Source: nature.com
Link: https://www.nature.com/articles/s41586-021-03819-2
Source snippet
Highly accurate protein structure prediction with AlphaFoldby J Jumper · 2021 · Cited by 49928 — The AlphaFold network directly predicts...
Source: arxiv.org
Link: https://arxiv.org/abs/2112.05527
Source: ebi.ac.uk
Title: strengths and limitations of alphafold
Link: https://www.ebi.ac.uk/training/online/courses/alphafold/an-introductory-guide-to-its-strengths-and-limitations/strengths-and-limitations-of-alphafold/
Source snippet
25 Jan 2024 — AlphaFold2 can be used to identify intrinsically disordered regions. Naturally, the system cannot predict disordered or dyn...
Source: sciencedirect.com
Link: https://www.sciencedirect.com/science/article/pii/S0022283621004411
Source snippet
AlphaFold and Implications for Intrinsically Disordered...by KM Ruff · 2021 · Cited by 714 — AlphaFold, a deep learning-based approach t...
Source: mdpi.com
Link: https://www.mdpi.com/2673-8856/5/4/46
Source snippet
Evaluating AlphaFold 3 Folding of the Intrinsically...by CM Nartey · 2025 — Background/Objectives: Intrinsically disordered protein regi...
Source: arxiv.org
Link: https://arxiv.org/abs/2510.15939
Source snippet
arXivHallucinations in AlphaFold3 for Intrinsically Disordered Proteins with disorder in Biological Process ResiduesOctober 8, 2025...

Published: October 8, 2025
Source: nature.com
Link: https://www.nature.com/articles/s41467-025-56572-9
Source snippet
NatureAlphaFold prediction of structural ensembles of disordered...by ZF Brotzakis · 2025 · Cited by 121 — We introduce the AlphaFold-Me...
Source: pmc.ncbi.nlm.nih.gov
Title: PMCAccurate Protein Dynamic Conformational Ensembles
Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC12469643/
Source snippet
Protein Dynamic Conformational Ensembles - PMCby D Lesovoy · 2025 · Cited by 2 — In this study, we present an approach that integrates fr...
Source: sciencedirect.com
Link: https://www.sciencedirect.com/science/article/abs/pii/S0165614714000509
Source snippet
ScienceDirectUnraveling structural mechanisms of allosteric drug actionby R Nussinov · 2014 · Cited by 141 — Our structural analysis reve...
Source: mdpi.com
Link: https://www.mdpi.com/1424-8247/19/3/480
Source snippet
AI-Driven Design of Miniproteins as Potential Allosteric...by X Liu · 2026 — Allosteric sites are often shallow, transient, or only popu...
Source: pnas.org
Link: https://www.pnas.org/doi/10.1073/pnas.2406407121
Source snippet
PNASAlphaFold-Multimer accurately captures interactions and...by A Omidi · 2024 · Cited by 75 — Our study demonstrates that AlphaFold-Mu...
Source: arxiv.org
Link: https://arxiv.org/abs/2311.15201
Source: sciencedirect.com
Link: https://www.sciencedirect.com/science/article/pii/S2950363925000286
Source snippet
ScienceDirectGeneration of appropriate protein structures for virtual...by Y Yasumitsu · 2025 · Cited by 6 — However, AlphaFold2 does no...
Source: sciencedirect.com
Link: https://www.sciencedirect.com/science/article/abs/pii/S0959440X25001332
Source snippet
Generation of protein dynamics by machine learningby G Janson · 2025 · Cited by 10 — This review examines emerging machine learning appro...
Source: arxiv.org
Link: https://arxiv.org/abs/2503.03989
Source: nature.com
Link: https://www.nature.com/articles/s41467-024-51507-2
Source snippet
Structure prediction of alternative protein conformationsby P Bryant · 2024 · Cited by 113 — Here, we train a structure prediction networ...
Source: nature.com
Link: https://www.nature.com/articles/s42003-026-10112-3
Source snippet
protein binder design and conformational state predictionby LA Abriata · 2026 — The communities of structural biology, molecular modeling...
Source: nature.com
Link: https://www.nature.com/articles/s41467-026-69172-y
Source snippet
Proc. Natl. Acad. Sci. USA 121...Read more...
Source: nature.com
Link: https://www.nature.com/articles/s41598-025-17022-0
Source snippet
A comprehensive application of FiveFold for conformation...by SK Niazi · 2025 · Cited by 2 — We discuss future applications in structure...
Source: nature.com
Link: https://www.nature.com/articles/s42004-025-01763-0
Source snippet
Challenging AlphaFold in predicting proteins with large-...by BH Perkins-Jechow · 2025 · Cited by 6 — Our analyses show that AlphaFold2...
Source: nature.com
Link: https://www.nature.com/articles/s41598-022-14382-9
Source snippet
AlphaFold2 models indicate that protein sequence...by HB Guo · 2022 · Cited by 303 — It had been shown that combining both the flexibili...
Source: arxiv.org
Link: https://arxiv.org/html/2508.18446v1
Source snippet
AlphaFold 3 as a Differentiable Framework for Structural...25 Aug 2025 — By treating protein folding and dynamics as a continuous, learn...
Source: arxiv.org
Link: https://arxiv.org/html/2503.03989v1
Source snippet
Integrating Protein Dynamics into Structure-Based Drug...6 Mar 2025 — Our method generates [meaningful]({{ 'ai-bloom-abun/ai-bloom-abun-98d3a6-ai-medicine-l-7c4177-clinical-seco-8e7b35-meaningful-cl-801d7d/' | relative_url &#12...
Source: sciencedirect.com
Link: https://www.sciencedirect.com/science/article/am/pii/S0010482524007054
Source snippet
Overview of AlphaFold2 and breakthroughs in overcoming...by L Wang · 2024 · Cited by 48 — Despite the limitations of AF2 in predicting d...
Source: mdpi.com
Link: https://www.mdpi.com/2674-0583/3/4/18
Source snippet
ChronoSort: Revealing Hidden Dynamics in AlphaFold3...by MJ Argyle · 2025 · Cited by 2 — This finding has important implications for dru...
Source: academic.oup.com
Link: https://academic.oup.com/pcm/article/8/3/pbaf015/8180385
Source snippet
3: an unprecedent opportunity for fundamental...by Z Fang · 2025 · Cited by 33 — With unprecedented accuracy, the AF3 model may successf...
Source: ebi.ac.uk
Link: https://www.ebi.ac.uk/training/online/courses/alphafold/alphafold-3-and-alphafold-server/introducing-alphafold-3/how-have-alphafold-3s-predictions-been-validated/
Source snippet
How have AlphaFold 3's predictions been validated?This study found that AF3 excels at predicting static protein-ligand interactions where...
Source: cell.com
Link: https://www.cell.com/cell-systems/fulltext/S2405-4712%2825%2900341-2
Source snippet
Proc. Natl. Acad. Sci. USA...Read more...
Source: elifesciences.org
Link: https://elifesciences.org/articles/88749
Source snippet
Drug Discovery: Decoding the mechanisms of allosteryby S Khan · 2023 · Cited by 3 — A complex interplay between structure, conformational...
Source: elifesciences.org
Title: reviewed preprints
Link: https://elifesciences.org/reviewed-preprints/94029
Source snippet
Reliable protein-protein docking with AlphaFold, Rosetta...by A Harmalkar · 2025 · Cited by 61 — In this work, we combine AlphaFold as...
Source: frontiersin.org
Link: https://www.frontiersin.org/journals/bioinformatics/articles/10.3389/fbinf.2022.959160/full
Source snippet
FrontiersImproving peptide-protein docking with AlphaFold-Multimer...by I Johansson-Åkhe · 2022 · Cited by 164 — We find that AlphaFold...
Source: elifesciences.org
Title: reviewed preprints
Link: https://elifesciences.org/reviewed-preprints/104901
Source snippet
eLifeHarnessing AlphaFold to reveal hERG channel...by K Ngo · 2025 · Cited by 13 — This valuable study uses AlphaFold2 to guide the stru...
Source: pmc.ncbi.nlm.nih.gov
Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC10844226/
Source snippet
PMC - NIHby W Lu · 2024 · Cited by 234 — AlphaFold, which leads the way in the prediction of nearly all structures in the human proteome...
Source: frontiersin.org
Link: https://www.frontiersin.org/journals/molecular-biosciences/articles/10.3389/fmolb.2022.1070328/full
Source snippet
Recent applications of computational methods to allosteric...by RG Govindaraj · 2023 · Cited by 26 — Here we examine the extent to which...
Source: science.org
Link: https://www.science.org/doi/10.1126/sciadv.adm9926
Source snippet
We demonstrate the power of the algorithm by generating variants of...Read mo...

Additional References

Source: medium.com
Link: https://medium.com/%40falk_hoffmann/promising-alternatives-to-alphafold-3-how-they-work-and-when-to-use-them-ccbcde490b14
Source snippet
Alternatives to AlphaFold 3— how they work and when to...AlphaFold 3 (AF3) sets the new standard for protein-ligand structure prediction...
Source: frontiersin.org
Link: https://www.frontiersin.org/research-topics/11378/understanding-protein-dynamics-binding-and-allostery-for-drug-design/magazine
Source snippet
Understanding Protein Dynamics, Binding and Allostery for...In this review, we discuss simulation-based multiscale approaches, experimen...
Source: deepmind.google
Link: https://deepmind.google/science/alphafold/
Source snippet
Google DeepMindAlphaFold — Google DeepMindAlphaFold has revealed millions of intricate 3D protein structures, and is helping scientists u...
Source: creative-biostructure.com
Link: https://www.creative-biostructure.com/integrating-alphafold-drug-discovery.htm?srsltid=AfmBOoq56SkjoXtTDft0PJhSpst7zbm659ziq95Ic8YrIuvmpMc2AgVg
Source snippet
Integrating AlphaFold into the Drug Discovery ProcessAlphaFold 3's unprecedented accuracy in predicting protein-ligand interactions allow...
Source: frontiersin.org
Link: https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1739303/full
Source snippet
The transformative impact of AI-enabled AlphaFold 3by C Chakraborty · Cited by 1 — Additionally, AlphaFold models are limited by their la...
Source: ouci.dntb.gov.ua
Link: https://ouci.dntb.gov.ua/works/4KrEABWN/
Source snippet
static structures: protein dynamic conformations...In this study, we present an approach that integrates free MD simulations, starting f...
Source: medium.com
Link: https://medium.com/%40cognidownunder/alphafold-changed-biology-forever-when-it-solved-protein-folding-78bb8768483a
Source snippet
AlphaFold 3 throws out the limitations entirely. Feed it a protein, some DNA, a drug candidate, and a few...Read more...
Source: youtube.com
Link: https://www.youtube.com/watch?v=7q8Uw3rmXyE
Source snippet
What Is AlphaFold? | NEJMAlpha fold compares it to several databases of protein sequences to extract similar sequences from various organ...
Source: inductive.bio
Title: strong baseline for alphafold 3 docking
Link: https://www.inductive.bio/blog/strong-baseline-for-alphafold-3-docking
Source snippet
Approaching AlphaFold 3 docking accuracy in 100 lines of...7 Jun 2024 — They find that AF3 has a 15% absolute improvement in PB-valid po...
Source: youtube.com
Link: https://www.youtube.com/watch?v=7NMhJQBdk7Q
Source snippet
Learning transferable protein backmapping from conformational ensembles...