Recep Adiyaman
bioinformatics

Issue #38: Evaluating zero-shot prediction of monomeric protein design success by AlphaFold, ESMFold, and ProteinMPNN.

February 02, 2026 Daily Intelligence
Protein Design Daily

Building something in Protein Design?

I love collaborating on new architectural challenges. Let's build together.

LinkedIn Connect Direct Message

🧬 Protein Design Digest

Curated protein signals by Recep Adiyaman

Join 1,000+ researchers. Unsubscribe anytime.

🚀 Today’s Top Signal

Evaluating zero-shot prediction of monomeric protein design success by AlphaFold, ESMFold, and ProteinMPNN.

🧬 Abstract

De novo protein design has enabled the creation of proteins with diverse functionalities that are not found in nature. Despite recent advances, experimental success rates remain inconsistent and context-dependent, posing a bottleneck for broader applications of de novo design. To overcome this, structure and sequence prediction models have been applied to assess design quality prior to experimental testing to save time and resources. In this study, we examined the extent to which AlphaFold, Protein MPNN, and ESMFold can discriminate between experimentally successful and unsuccessful designs. We first curated a benchmark dataset of 614 experimentally characterized de novo designed monomers from 11 different design studies between 2012 and 2021. All predictive models demonstrated moderate ability to discriminate experimental successes (expressed, soluble, monomeric, and fold with the correct secondary structure) from failures. Still, many failed designs have better confidence metrics than successful designs, and confidence metrics were topology-dependent. Among all computational models evaluated, ESMFold average predicted local-distance difference test (pLDDT) yielded the best individual performance at distinguishing between successful and unsuccessful designs. A logistic regression model combining all confidence metrics provided only modest improvement over ESMFold pLDDT alone. Overall, these results show that these models can serve as an initial filtering strategy prior to experimental validation; however, their utility at accurately predicting experimentally successful designs remains limited without task-specific training.

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

⭐ Additional Signals

Comparison of In Vitro Multiple Physiological Activities of Cys-Tyr-Gly-Ser-Arg (CYGSR) Linear and Cyclic Peptides and Analysis Based on Molecular Docking.

Peptide cyclization is a strategy to improve biological stability and functional activity, but direct comparison between linear and cyclic peptides with the same sequence is still limited. In this study, linear (L-CR5) and cyclic (C-CR5) forms were synthesized, and biological functions such as antioxidant, whitening, and anti-wrinkle activity were compared and evaluated. C-CR5 showed about 22.3 times of DPPH radical scavenging activity, which was significantly stronger than L-CR5, and tyrosinase inhibition increased rapidly in C-CR5 to reach inhibition of 95% or more, whereas L-CR5 showed only moderate activity in the same range (about 6.5 times). MMP-1 expression in the evaluation of anti-wrinkle activity did not show a decreasing trend in L-CR5 at all, while C-CR5 showed an anti-wrinkle effect, which was reduced by about 92.8% at 400 μg/mL. As a result of molecular docking analysis, C-CR5 exhibited lower MolDock scores than L-CR5 toward both tyrosinase and MMP-1, indicating a potentially higher binding affinity and improved binding stability. This is expected to be due to reduced structural flexibility and optimized residue directions (especially Tyr and Arg). These results indicate that peptide cyclization is an example of enhanced functional bioactivity of CYGSR and provides a positive case for the structure-activity relationship.

Decrypting potential mechanisms linking ochratoxin A to hepatocellular carcinoma: an integrated approach combining toxicology, machine learning, molecular docking, and molecular dynamics simulation.

Background Ochratoxin A (OTA), a common food-borne mycotoxin, is a potential human carcinogen, yet the specific molecular mechanisms linking it to hepatocellular carcinoma (HCC) remain unclear. Methods We integrated network toxicology to predict OTA targets and intersected them with HCC transcriptomic data to identify key candidate genes. Functional enrichment analysis was then conducted. Multiple machine learning algorithms were applied to screen and validate core genes. Furthermore, molecular docking and molecular dynamics (MD) simulations were employed to evaluate the binding stability between OTA and key target proteins. Results A total of 50 key genes were identified as potential targets for potential OTA-associated hepatocarcinogenesis. Enrichment analysis revealed their significant involvement in critical processes such as xenobiotic metabolism and oxidative stress response. Machine learning analysis prioritized eight core genes (AURKA, GABARAPL1, CA2, PARP1, LMNA, SLC27A5, EPHX2, and GSTP1), and a combined diagnostic model demonstrated outstanding performance (AUC = 0.986). Structural analyses via molecular docking and MD simulations confirmed stable binding interactions between OTA and these core targets. Conclusions This integrated computational study identifies a set of candidate genes through which OTA may potentially interact with HCC-associated molecular networks. The robust binding predicted between OTA and the core targets provides a structural basis for these interactions. These findings offer a prioritized list of targets and a theoretical framework for subsequent experimental validation and investigation into OTA’s toxicological role in HCC.

Study on the Mechanism of Ku Diding in the Treatment of Diabetes based on Network Pharmacology, Molecular Docking Technology, and Molecular Dynamics.

Introduction To explore how Ku Diding (KDD) works in managing Diabetes Mellitus (DM), researchers utilized network pharmacology, molecular docking, and molecular dynamics methodologies. Methods Key active components of KDD were identified using the Traditional Chinese Medicine Systematic Pharmacology Database and Analysis Platform (TCMSP). Data for diabetesrelated targets were retrieved from the Human Genetic Comprehensive Databases (Genecards) and the Online Mendelian Inheritance in Man (OMIM) database. The intersection of these targets was analyzed to determine potential therapeutic targets for diabetes treatment. Proteinprotein interaction networks (PPI) were constructed using the STRING database and Cytoscape software, followed by Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. Molecular docking between the components and key targets was performed using the AutoDock Vina platform. Results This study identified that Dihydrosanguinarine, (S)-Scoulerine, among others, are the main active ingredients of KDD for treating DM, showing high affinity for critical targets like PTGS2 and PRKACA, through multiple pathways including vascular regulation, neuromodulation, metabolic regulation, and endocrine regulation. The molecular docking results showed that there are interactions between the active ingredients and the key targets, with the majority of the effective components exhibiting a stronger binding affinity than Metformin. Among them, (S)-Scoulerine and Dihydrosanguinarine demonstrated high docking affinity with the key target proteins PTGS2 and PRKACA. Discussion DM is closely linked to oxidative stress, chronic inflammation, and insulin signaling dysregulation. This study reveals that KDD exerts anti-diabetic effects via a multi-target network involving proteins such as PRKACA, PTGS2, ESR1, FOS, and DRD2. These targets are associated with glucose metabolism, inflammation, oxidative stress, and neural regulation. Modulation of these pathways likely enhances insulin sensitivity, lowers blood glucose, suppresses inflammation, and protects against oxidative damage. GO and KEGG analyses further indicate involvement in MAPK signaling, synaptic transmission, and vascular regulation, forming a multidimensional “metabolism-inflammation-neural” regulatory network. Compared to Metformin, most KDD-derived compounds showed stronger binding, highlighting their therapeutic potential. Molecular dynamics simulations support the stability of the observed binding conformations, suggesting their potential as therapeutic targets. These findings underscore KDD’s ability to simultaneously target multiple pathological mechanisms, offering a holistic treatment strategy for DM. Conclusion This study provides preliminary evidence that KDD is characterized by a multicomponent, multi-target, and multi-pathway approach in the treatment of diabetes mellitus (DM), thereby establishing a scientific foundation for further in-depth exploration of KDD’s molecular mechanisms.

🧪 AI & Research News

🏢 Industry Insight & Applications

⚡ Quick Reads

Vision-Language Controlled Deep Unfolding for Joint Medical Image Restoration and Segmentation

We propose VL-DUN, a principled framework for joint All-in-One Medical Image Restoration and Segmentation (AiOMIRS) that bridges the gap between low-level signal recovery and high-level semantic understanding. While standard pipelines treat these tasks in isolation, our core insight is that they are fundamentally synergistic: restoration provides clean anatomical structures to improve segmentation, while semantic priors regularize the restoration process. VL-DUN resolves the sub-optimality of sequential processing through two primary innovations. (1) We formulate AiOMIRS as a unified optimization problem, deriving an interpretable joint unfolding mechanism where restoration and segmentation are mathematically coupled for mutual refinement. (2) We introduce a frequency-aware Mamba mechanism to capture long-range dependencies for global segmentation while preserving the high-frequency textures necessary for restoration. This allows for efficient global context modeling with linear complexity, effectively mitigating the spectral bias of standard architectures. As a pioneering work in the AiOMIRS task, VL-DUN establishes a new state-of-the-art across multi-modal benchmarks, improving PSNR by 0.92 dB and the Dice coefficient by 9.76%. Our results demonstrate that joint collaborative learning offers a superior, more robust solution for complex clinical workflows compared to isolated task processing. The codes are provided in https://github.com/cipi666/VLDUN.

Computational Screening, ADME Study, and Evaluation of Benzothiazole Derivatives as Potential Anticancer Agents.

Benzothiazole derivatives have garnered considerable interest in medicinal chemistry due to their diverse biological activities, including anticancer potential. The synthesis of 2-substituted benzothiazoles is traditionally achieved via two main approaches: (1) condensation of 2-aminothiophenols with aldehydes or carboxylic acid derivatives under highly acidic conditions, and (2) cyclization of thiobenzanilides. In this study, approximately 65 benzothiazole analogs were evaluated for anticancer potential using in silico tools and ADME profiling. ADME properties were predicted using SwissADME, while molecular docking studies were performed using Molegro Virtual Docker 6.0. Gefitinib and Erlotinib were used as reference drugs for both pharmacokinetic and in silico comparisons. Biological activity predictions were conducted using the PASS online web server. Docking scores for the analogs ranged from -134.60 to -114.36, with several compounds outperforming standard drugs Gefitinib (-122.87) and Erlotinib (-119.22). Compounds 12, 17, 27, 43, and 49 exhibited five hydrogen bond interactions, whereas compound 45 showed a maximum of six, exceeding the interactions observed for the standard drugs. Most compounds had molecular weights below 500 and favorable Log P values (e.g., compounds 4: 2.34, 5: 2.85, 7: 2.56, 10: 2.76, 17: 2.78, 19: 2.51, 26: 2.09, 30: 1.20, 40: 1.78, 45: 1.76, 56: 1.75), lower than the reference drugs (3.92, 3.20). Selected compounds also displayed improved topological polar surface area (TPSA) values (e.g., 5: 80.05 Ų, 11: 79.46 Ų, 13: 71.83 Ų, 15: 87.74 Ų, 23: 68.82 Ų, 32: 61.36 Ų, 36: 45.53 Ų, 52: 41.13 Ų) compared to standard drugs (68.74 Ų, 74.73 Ų). Targeting EGFR using PASS predictions, compounds 32, 33, 35, 39, 46, and 48 exhibited activities similar to Gefitinib and Erlotinib. Docking and ADME analyses indicated that several benzothiazole analogs outperformed standard drugs in binding affinity and pharmacokinetic profiles. EGFR, a transmembrane receptor tyrosine kinase, plays a central role in cell proliferation, survival, angiogenesis, and migration. Most compounds demonstrated good gastrointestinal absorption, suggesting favorable oral bioavailability according to Lipinski, Ghose, Veber, Egan, and Muegge rules. PASS predictions indicated potential anticancer activities, including inhibition of transcription factor STAT3, DNAdirected RNA polymerase, Mcl-1, proto-oncogene tyrosine-protein kinase Fgr, and EGFR, with potential antineoplastic effects across multiple cancer types, including solid tumors, lung, gastric, lymphoma, sarcoma, breast, and pancreatic cancers. Compounds 12, 17, 27, 43, 45, and 49 demonstrated strong binding affinities and superior pharmacokinetic profiles compared to Gefitinib and Erlotinib. Overall, benzothiazole derivatives represent a promising scaffold for the design of EGFR inhibitors, potentially contributing to targeted anticancer therapy.

Novel and sustainable microfabricated Cu ion selective sensor doped with ionophore and supported with docking study for determination of vonoprazan fumarate in tablet dosage form.

In the field of drug analysis, there is a growing emphasis on developing techniques that are environmentally friendly, cost-effective, and efficient. To align with the principles of green analytical chemistry and to support the advancement of portable and handheld devices, an innovative microfabricated ion selective electrode (ISE) has been developed for the detection of Vonoprazan fumarate (VON). The development of this electrode involved a two-step optimization process. Initially, a range of ionophores were screened to determine the one with the highest selectivity for VON. Through molecular docking studies, gamma-cyclodextrin (γ-cyclodextrin) was identified as demonstrating maximal activity towards VON. The second optimization step involved incorporating a graphene nanocomposite as an ion to electron transducer layer between the γ-cyclodextrin polymeric membrane and the microfabricated copper (Cu) solid contact ISE. This nanocomposite layer contributed to enhanced stability, reduced potential drift, and rapid response times (approximately 30 s), likely due to its hydrophobic properties that prevent water layer formation at the interface between the Cu electrode and the polymeric membrane. The VON sensor was characterized according to IUPAC guidelines, revealing a linear dynamic range of 2.00 × 10⁻5 to 1.00 × 10⁻2 M (equivalent to 9.23 to 4615.00 µg/mL) and a limit of detection (LOD) of 1.00 × 10⁻5 M. This sensor was successfully utilized for the selective determination of VON in bulk powder and pharmaceutical formulations. Statistical analysis showed no significant difference when comparing the results with those obtained using the reported method. The environmental impact of the method was assessed using Complex-GAPI and BAGI tools.

ME-PFP: An Ensemble Learning Approach Fusing Multi-Source Features for Protein Function Prediction.

Proteins, as essential components of living organisms, play a critical role in both drug discovery and disease mechanism research. Multiple empirical studies have shown that there is a significant correlation between protein function and drug targets with therapeutic potential. Therefore, how to accurately and efficiently predict protein function is an urgent issue that needs to be addressed. Existing research faces challenges such as insufficient utilization of protein data and low heterogeneous fusion performance. In this paper, we propose ME-PFP, a novel ensemble learning framework that integrates sequence representations from a protein language model, domain, and protein-protein interaction data to improve protein function prediction. To effectively capture and utilize heterogeneous features, we design three specialized attention-based feature extractors tailored to each data modality. These features are then fused through a dynamic weighting strategy to enable complementary information exchange between different modalities, thereby improving protein function prediction performance. Extensive experiments on benchmark data sets show that ME-PFP significantly outperforms sequence-based and multisource fusion models. Notably, it achieved an average improvement of 13.23% on the human data set and 11.11% on the yeast data set. The experimental results show that this study not only improves the accuracy of protein function prediction, but also promotes progress in the field of computational biology.

Impact of clove oil on behavioral and biochemical parameters in restrained rats by using docking and experimental approaches.

Background Acute restraint stress activates the (HPA) axis, elevating corticosterone and influencing cognitive function. Clove oil (Syzygium aromaticum), known for its antioxidant and neuroprotective properties, may counteract stress-induced biochemical and behavioral alterations. Objective This study evaluated the effects of clove oil pretreatment on stress-induced memory changes and biochemical responses in rats, supported by molecular docking of its active constituents. Methods Rats were divided into stressed and unstressed groups. Memory performance was assessed using the Morris water maze (MWM) for long-term memory and an elevated plus maze (EPM) for short-term memory. Plasma corticosterone levels and acetyl cholinesterase (AChE) activity were measured. Molecular docking was performed to assess interactions between clove oil constituents and AChE. Results Acute restraint stress (2 hours) significantly enhanced long-term and short term memory (p Conclusion Clove oil exhibits neuroprotective and cognition-enhancing effects in stress-exposed rats, suggesting its potential therapeutic value for managing stress-related cognitive impairments.

Enhancing Drug Repurposing with Consensus Docking: Discovery of Novel Discoidin Domain Receptor 1 Inhibitors.

Developing novel drugs is a long and difficult process, particularly in oncology, where high attrition rates make clinical trials costly and time-consuming. In response, drug repurposing emerges as an efficient alternative: existing compounds can be used effectively in new therapeutic contexts. Discoidin domain receptor 1 (DDR1), a receptor tyrosine kinase involved in tumor progression, has emerged as a promising target for solid tumor treatment. To identify novel potential DDR1 inhibitors, we applied ESSENCE-Dock, an in-house consensus docking method designed to improve hit enrichment. Using this approach, we performed virtual screening of the DrugBank database, a comprehensive collection of Food and Drug Administration-approved and investigational drugs. To ensure novelty, we performed fingerprint dissimilarity analysis against known DDR1 inhibitors from BindingDB, prioritizing structurally distinct candidates. While several known DDR1 inhibitors ranked among the top-scoring compounds, we prioritized novel, structurally distinct candidates for testing. Biochemical IC50 assays validated three previously unreported DDR1 inhibitors with nanomolar potency, while molecular dynamics simulations confirmed their stable binding within the DDR1 active site. Functional cell-based assays revealed inhibition of DDR1-mediated signaling and cancer cell migration. These findings demonstrate the effectiveness of our consensus-based virtual screening approach in drug repurposing and underscore its potential to streamline oncology drug development.

Regarding “The molecular mechanisms through which psilocybin prevents suicide: evidence from network pharmacology and molecular docking analyses”.

Enzymatic preparation of epigallocatechin gallate diglucoside and bioactivity assessment.

Epigallocatechin gallate (EGCG), the predominant catechin in green tea, has limited application due to its poor water solubility and instability. To address these issues, this study utilized recombinant sucrose phosphorylase to catalyze the glucosylation of EGCG, successfully synthesizing (-)-epigallocatechin gallate 4’,4″-O-α-D-diglucopyranoside (EGCG-2G). The process was optimized using response surface methodology, achieving a 97.46% conversion rate of EGCG. EGCG-2G was purified to ≥ 99% purity by semi-preparative liquid chromatography. It exhibited approximately 124-fold higher water solubility than EGCG and demonstrated significantly enhanced stability under thermal and acidic conditions (50 °C, pH = 5), with an 84.82% and 35.36% improvement over EGCG and commercial (-)-epigallocatechin-3-gallate-4’-O-α-D-glucoside (EGCG-1G), respectively. Furthermore, EGCG-2G displayed notable antioxidant, anti-inflammatory, and anti-melanogenic activities. It effectively scavenged intracellular and extracellular free radicals, reduced inflammatory cytokine levels, and inhibited melanin synthesis. Molecular docking and gene expression analyses suggested that its anti-melanogenic effect might be associated with the MC1R/cAMP/MITF signaling pathway.

💡 Pipeline Tip

Employ HADDOCK for ambiguous restraints in protein-protein docking.

🛠️ Resources

Deep learning is not a magic wand, but a powerful lens for structural biology. — Recep Adiyaman

BS HF DK