Issue #120: Adversarial Sequence Mutations in AlphaFold and ESMFold Reveal Nonphysical Structural Invariance, Confidence Failures, and Concerns for Protein Design

Signal of the Day

Adversarial Sequence Mutations in AlphaFold and ESMFold Reveal Nonphysical Structural Invariance, Confidence Failures, and Concerns for Protein Design

AlphaFold has transformed structural biology and spawned an ecosystem of derivative tools for protein design, binding prediction, and drug discovery. However, whether AlphaFold has learned generalizable biophysical principles versus template-based pattern matching remains unclear—a distinction critical for applications beyond its training context. Here, we perform a systematic adversarial evaluation of AlphaFold 3 using point and deletion mutations across 200 proteins. Remarkably, predicted structures remain invariant to mutations of up to 40% of residues—including deliberately destabilizing substitutions—and to deletions of 10%. Notably, this invariance holds even for experimentally validated fold-switching proteins that are known to adopt alternative conformations in response to such mutations, despite the fact that these proteins are small and monomeric—precisely the category where AlphaFold is expected to perform best. Confidence metrics prove unreliable, as they select the most accurate structure at most 35% of the time and correlate with the structural quality of the best available training set template. This suggests that AlphaFold’s uncertainty estimates reflect template availability more than biophysical reasoning. ESMFold exhibits greater, though still imperfect, mutational sensitivity, suggesting superior sequence-structure coupling. These findings indicate that AlphaFold may rely heavily on memorized templates rather than biophysical reasoning, with profound implications for the reliability of AlphaFold-based protein design, drug discovery, and modeling workflows.

Why this matters: Critical for improving fold accuracy and reducing structural uncertainty in de novo design.

Also Worth Reading

Tocotrienol as a multi-target inhibitor of ICAM-1, VCAM-1, and E-selectin: Comparison using AutoDock and GNINA docking with molecular dynamics simulation.

Atherosclerosis is a chronic inflammatory disease characterized by endothelial dysfunction and leukocyte adhesion, mediated by cell adhesion molecules such as E-selectin, intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1). Tocotrienols, a subgroup of vitamin E, exhibit potent antioxidant and anti-inflammatory properties, suggesting their potential role in attenuating atherosclerosis. This study comparatively evaluated the binding affinities and molecular interaction profiles of α-, β-, γ-, and δ tocotrienol isomers towards E-selectin, ICAM-1, and VCAM-1 using molecular docking approaches, followed by molecular dynamic simulation to assess the stability of the top-ranked protein-ligand complexes. The docking experiment was conducted using MolModa, an automated molecular docking platform based on AutoDock Vina and convolutional neuronal network (CNN)-based AI-assisted GNINA. Overall, the conventional molecular docking tool AutoDock Vina results showed that all tocotrienol isomers exhibited the strongest average binding affinities to VCAM-1. Among the isomers, α-tocotrienol displayed the highest binding affinity towards E-selectin (-6.69 ± 0.00 kcal/mol) and ICAM-1 (-6.79 ± 0.00 kcal/mol), whereas β-tocotrienol exhibited the strongest affinity toward VCAM-1 (-7.59 ± 0.00 kcal/mol) in the molecular docking analysis using conventional molecular docking tool AutoDock Vina. In contrast, the AI-assisted molecular docking tool GNINA leveraging deep learning, demonstrated a more accurate and consistent affinity profile by consistently identified β-tocotrienol as the most favorable binder toward E-selectin (-6.91 ± 0.01 kcal/mol) and ICAM-1 (-7.08 ± 0.90 kcal/mol), characterized by hydrogen bonding, hydrophobic interactions, and extensive van der Waals forces, that are crucial for the lipid-soluble ligand. The AI-assisted molecular docking tool GNINA docking for VCAM-1 was not generated due to structural limitations of the receptor model. Molecular dynamics (MD) simulations over 200 ns demonstrate a significant stabilizing interaction with GLU87, whereas the hydrogen bonding at ASP178 was found to be intermittent and contributory throughout the trajectory. This study provides the first comprehensive computational evidence differentiating the multi-target potency of tocotrienol isomers in targeting inflammatory and vascular-related pathways. Further experimental validation is warranted to confirm these in silico predictions and explore their biological significance.

Exploring the mechanism of saffron in treating viral myocarditis using network pharmacology and molecular docking.

Viral myocarditis (VM) is a cardiovascular disorder that can lead to heart failure and cardiogenic shock. Saffron, a traditional Chinese medicinal herb, has shown therapeutic potential against VM in numerous studies. However, the mechanisms through which saffron exerts its effects on VM remain poorly understood. Thus, this study aimed to elucidate the active compounds, molecular targets, and signaling pathways involved in saffron’s therapeutic action against VM by employing network pharmacology and molecular docking approaches. The active compounds and corresponding targets of saffron were retrieved from the Traditional Chinese Medicine Systems Pharmacology database. VM-associated targets were sourced from the GeneCards database. Overlapping targets between saffron and VM were then identified. Protein-protein interaction networks were established and analyzed utilizing the STRING platform and Cytoscape software to determine core targets. Furthermore, gene ontology and Kyoto encyclopedia of genes and genomes enrichment analyses were carried out utilizing Bioconductor in R to explore the potential biological activities and signaling pathways through which saffron may act against VM. Finally, molecular docking and model visualization were carried out using AutoDock Tools and PyMOL open-source software. From the database, we identified 4 active compounds in saffron with potential effects against VM: crocetin, isorhamnetin, kaempferol, and quercetin. A total of 60 corresponding targets were observed, with TNF, IL-6, IL-1β, CXCL8, and JUN emerging as core targets. Kyoto encyclopedia of genes and genomes enrichment analysis revealed 155 regulatory signaling pathways, among which the TNF, AGE-RAGE, and IL-17 signaling pathways, lipid metabolism, and atherosclerosis were the most prominent. Molecular docking results indicated that quercetin showed the strongest binding affinity toward IL-1β and CXCL8. The therapeutic effect of saffron against VM is not driven by a single factor, but rather involves multiple active compounds, targets, and signaling pathways.

InversePep: Diffusion-driven structure-based inverse folding for functional peptides.

Designing functional peptides with specific structural and biochemical properties is critical for applications in protein engineering and therapeutic discovery. However, most peptide design approaches rely on evolutionary or local sequence optimization methods, which are limited when adapting to peptides’ shorter length, high conformational flexibility, and unique physicochemical constraints. While recent structure-based inverse folding models have shown success for proteins, these models often underperform on peptides because sequence recovery alone is not a reliable indicator of stability or foldability in short, flexible backbones. To address this challenge, we introduce InversePep, a generative diffusion model for structure-based peptide inverse folding. InversePep learns the conditional distribution of sequences that can adopt a given backbone conformation, enabling direct generation of peptides tailored to target structural geometries. The framework integrates a geometric graph neural network to encode 3D backbone features with a Transformer-based sequence refinement module that iteratively denoises candidate sequences during diffusion. Trained on a diverse set of peptide backbones sourced from Propedia and SATPdb, InversePep effectively captures structural and biochemical diversity across peptide families. In systematic evaluations on held-out peptide structures and the PepBDB benchmark dataset, InversePep achieves Mean TM-SCORE (0.51), Median TM-SCORE (0.483), Mean RMSD-Simple (1.02), Median RMSD-Simple (0.97), Mean RMSD-Common (3.13), Median RMSD-Common (2.16), outperforming ProteinMPNN, and ESM-IF1 in generating geometry-consistent peptide sequences. In-silico folding analyses confirm that sampled peptides reliably adopt the target conformations. These results highlight InversePep’s capability for designing structurally stable and sequence-diverse peptides, demonstrating its potential in antimicrobial peptide discovery, peptide therapeutics, and molecular probe development.

Research & AI Updates

• Vertu’s Alphafold foldable phone has launched, and you’ve got to see its obnoxious price - MSN — Vertu’s Alphafold foldable phone has launched, and you’ve got to see its obnoxious price    MSN.

From the Industry

• Molecular Partners Holds Presentations at ASCO and SNMMI 2026 on DLL3-Targeting Radio-DARPin MP0712, Now Dosing Patients in Phase 1 - The Manila Times — Molecular Partners Holds Presentations at ASCO and SNMMI 2026 on DLL3-Targeting Radio-DARPin MP0712, Now Dosing Patients in Phase 1    The Manila Times.
• Why did WOK stock skyrocket nearly 300% in just 2 days? - MSN — Why did WOK stock skyrocket nearly 300% in just 2 days?    MSN.
• Rapid Coomassie Stains Market Forecast Points Higher Toward 2035, Driven by Biomanufacturing QC Expansion - IndexBox — Rapid Coomassie Stains Market Forecast Points Higher Toward 2035, Driven by Biomanufacturing QC Expansion    IndexBox.
• Pfizer Inks Up to $10.5B Innovent Biotech Licensing Deal - Gotrade — Pfizer Inks Up to $10.5B Innovent Biotech Licensing Deal    Gotrade.
• Fierce Pharma Asia—More China biotech hawkishness; Pfizer’s $10B Innovent deal; Astellas’ roadmap - Fierce Pharma — Fierce Pharma Asia—More China biotech hawkishness; Pfizer’s $10B Innovent deal; Astellas’ roadmap    Fierce Pharma.
• China’s Innovent Biologics signs US$10.5b Pfizer deal for 12 cancer drug trials - South China Morning Post — China’s Innovent Biologics signs US$10.5b Pfizer deal for 12 cancer drug trials    South China Morning Post.
• Innovent Biologics, Pfizer strike $10.5 bln cancer drug deal amid China biotech boom - Reuters — Innovent Biologics, Pfizer strike $10.5 bln cancer drug deal amid China biotech boom    Reuters.

Quick Reads

Design, Microwave-Assisted Synthesis, and Biological Evaluation of Novel 1,3-Thiazolidin-4-One Hybrids: Insights From DFT, Molecular Docking, and ADMET Profiling.

A highly efficient, environmentally benign one-pot multicomponent protocol was developed for synthesizing novel 1,3-thiazolidin-4-one derivatives via microwave irradiation of 3-pyridyl isothiocyanate, primary amines, and ethyl chloroacetate. Read more →

Integrated DFT, molecular docking, and molecular dynamics investigation of some novel 2-thiohydantoin analogues as potent CDK2 inhibitors for anticancer therapy.

Cancer progression is driven by dysregulation of cyclin-dependent kinase 2 (CDK2), a critical cell cycle regulator. Read more →

Kaempferol alleviates copper-induced nephrotoxicity by targeting MAPK8 and CASP3 in broiler: Insights from network toxicology, network pharmacology, molecular docking and dynamics simulation.

Copper (Cu) exposure poses significant threats to renal health yet effective mitigation strategies remain scarce. Read more →

Integrated Pharmacophore Modeling, Molecular Docking, and Molecular Dynamics Simulations Accelerate the Discovery of Novel PDE1 Inhibitors with Potential for the Treatment of Idiopathic Pulmonary Fibrosis

Conformational dynamics and inhibition mechanism of cyclin-dependent kinase 2: Insights from molecular docking and dynamics simulations with pyrazolopyrimidine analogues of roscovitine.

Cyclin-dependent kinase 2 (CDK2) is a key regulatory protein, controlling cell cycle progression through its conformational and catalytic mechanisms. Read more →

Computationally driven discovery of thieno-pyridine thiadiazoles: a synthesis, DFT, docking and dynamics roadmap toward multitarget antidiabetic therapy.

Diabetes mellitus remains a major metabolic disorder requiring new enzyme-targeted therapies. Read more →

Integrated LC-MS/MS profiling, network pharmacology, molecular docking, and enzyme inhibition assays reveal the antidiabetic potential of Mitragyna speciosa.

Mitragyna speciosa has gained attention as a medicinal plant with potential metabolic benefits, yet its antidiabetic mechanism remains unclear. Read more →

Genomic Analysis and Antimicrobial Resistance Profiling of Klebsiella pneumoniae Clinical Isolates from Urinary Tract Infections in India: Prevalence of Class A β-lactamase with Molecular Docking and Simulation Insights

Urinary tract infections (UTIs) are a major global health concern, with Klebsiella pneumoniae emerging as a prominent multidrug-resistant uropathogen. Read more →

Pipeline Tip

Check for missing residues in PDB files using PDB-Fixer before simulation.

Resources & Tools

• Dataset: UniRef - Clustered protein sequence sets for fast similarity searches.
• Dataset: BFD - Big Fantastic Database for deep learning protein modeling.
• Tool: MMseqs2 - Fast and sensitive sequence search and clustering suite. View all tools →
• Tool: HHSuite - Remote homology detection with HMM-HMM comparison. View all tools →
• Event: Protein Design Hub (LinkedIn Group) (Ongoing)
• Event: Structural Biology Events (Open)
• Job: MSc for Brain Tumor Immunology - Academic Positions at Academic Positions
• Job: PhD for Tumor Immunology and Immunotherapy - Academic Positions at Academic Positions

The protein structure is the language of life; design is its poetry. — Recep Adiyaman