Issue #3: A Comparative Study of Deep Learning and Classical Modeling Approaches for Protein-Ligand Binding Pose and Affinity Prediction in Coronavirus Main Proteases.

Signal of the Day

A Comparative Study of Deep Learning and Classical Modeling Approaches for Protein-Ligand Binding Pose and Affinity Prediction in Coronavirus Main Proteases.

The accurate prediction of protein-ligand binding poses and affinities is central to structure-based drug design. In this study, we first benchmarked three distinct pose generation strategies for data sets from the ASAP Antiviral Challenge 2025: molecular docking (Glide and AutoDock Vina), ligand-based superposition (FlexS), and deep learning-based modeling (AlphaFold3, Boltz-2, DiffDock and Gnina). We evaluated their performance on binding pose prediction for ligands targeting SARS-CoV-2 and MERS-CoV main protease (Mpro). For binding affinity estimation, we implemented a machine learning-based scoring approach called ligand-residue interaction profile scoring function (LRIP-SF), which integrates molecular mechanics generalized Born surface area (MM-GBSA) energy decomposition with machine learning algorithms. Our results showed that deep learning-based modeling with AlphaFold3 achieved the highest pose prediction accuracy with a success rate of 88.1% and an average ligand root-mean-square deviation (LRMSD) of 1.12 Å. Moreover, binding poses predicted by AlphaFold3 enabled the most accurate potency predictions by LRIP-SF, with the lowest mean absolute error (MAE) and root-mean-square error (RMSE) in pIC50 units across both targets: the MAE and RMSE are 0.606 and 0.813, respectively, for MERS-CoV Mpro and 0.724 and 0.894 respectively for SARS-CoV-2 Mpro. Although ligand-based superposition method (FlexS) was less accurate in pose prediction, it offered competitive potency prediction performance with significantly lower computational cost. To interpret model predictions by LRIP-SF and identify critical binding determinants, we performed global sensitivity analysis (GSA), revealing key residues that contributed most significantly to ligand binding. These findings highlight the importance of pose quality and interaction profiling in affinity prediction and demonstrate the great potential of deep learning-based methods for drug discovery, especially in the absence of cocrystal structures.

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

Also Worth Reading

Cytotoxicity, apoptosis, molecular docking, and molecular dynamics study of novel compounds of Sulfamide derivatives coupled with DHP scaffolds as potent inhibitors of the MCF-7, A549, SKOV-3, and EA. yh926 carcinoma cells.

A novel series of dihydropyridine-sulfonyl derivatives (AG-CHO and analogues A1-A7) were synthesized and structurally characterized. Molecular docking demonstrated favorable binding of these compounds to autophagy-associated and cancer-related targets, while molecular dynamics simulations confirmed A5 as the most stable ligand protein interactions. Functional assays in SKOV-3, MCF-7, A549, and EA.hy.926 cells using acridine orange staining and flow cytometry revealed significant autophagy induction. Among all tested compounds AG-CHO emerged as the most potent inducer of autophagy. Notably, derivatives such as A6 and A7 showed selective potency in endothelial cells, whereas A1, A5, and A7 were effective in A549 cells, indicating cell-specific activity. Collectively, this integrated computational and experimental study identifies A5 as the lead compound and highlights dihydropyridine-sulfonyl scaffolds as promising autophagy modulators and potential anticancer candidates for further preclinical development.

Meeko: Molecule Parametrization and Software Interoperability for Docking and Beyond.

Molecule parametrization is an essential requirement to guarantee the accuracy of docking calculations. Parametrization includes a proper perception of chemical properties such as bonds, formal charges and protonation states. This includes large biological macromolecules, such as proteins and nucleic acids, and small molecules, such as ligands and cofactors. The structures of proteins and nucleic acids are challenging due to omission of several atoms from the structural model, and from the lack of connectivity and bond order information in the PDB and mmCIF file formats. For small molecules, the very large chemical diversity poses challenges for both validating correctness and providing accurate parameters. These challenges affect various modeling approaches like molecular docking and molecular dynamics. Moreover, several specialized methods (particularly in molecular docking) leverage specific chemical properties to add custom potentials, pseudoatoms, or manipulate atomic connectivity. To address these challenges, we developed Meeko, a molecular parametrization Python package that leverages the widely used RDKit cheminformatics library for a chemically accurate description of the molecular representation. Small molecules are modeled as single RDKit molecules, and biological macromolecules as multiple RDKit molecules, one for each residue. Meeko is highly customizable and designed to be easily scriptable for high-throughput processing, replacing MGLTools for receptor and ligand preparation.

Research & AI Updates

• DFG announces 2026 Gottfried Wilhelm Leibniz Prize recipients - Philanthropy News Digest — DFG announces 2026 Gottfried Wilhelm Leibniz Prize recipients    Philanthropy News Digest.
• Structural Findings Reveal How Distinct GPCR Ligands Create Different Levels of Activation | Newswise - Newswise — Structural Findings Reveal How Distinct GPCR Ligands Create Different Levels of Activation | Newswise    Newswise.
• Largest protein classification in history finds 700,000 unknown structures - Earth.com — Largest protein classification in history finds 700,000 unknown structures    Earth.com.
• Kolmogorov-Arnold networks bridge AI and scientific discovery by increasing interpretability - Phys.org — Kolmogorov-Arnold networks bridge AI and scientific discovery by increasing interpretability    Phys.org.
• Bets on Generative AI to Redefine Drug Discovery——IntelliGenAI and their foundation model approach - Pandaily — Bets on Generative AI to Redefine Drug Discovery——IntelliGenAI and their foundation model approach    Pandaily.

From the Industry

• Layoff Tracker: Voyager Loses 30 Employees After Novartis Prunes Deal - BioSpace — Layoff Tracker: Voyager Loses 30 Employees After Novartis Prunes Deal    BioSpace.
• Galux, Boehringer Ingelheim to Jointly Explore AI in Precision Protein Design - Contract Pharma — Galux, Boehringer Ingelheim to Jointly Explore AI in Precision Protein Design    Contract Pharma.
• Profluent Bio Partners with Ensoma for AI-Designed Base Editors in Stem Cell Therapies - SynBioBeta — Profluent Bio Partners with Ensoma for AI-Designed Base Editors in Stem Cell Therapies    SynBioBeta.
• Europe Protein Engineering Market Size & Share, 2033 - Market Data Forecast — Europe Protein Engineering Market Size & Share, 2033    Market Data Forecast.

Quick Reads

A fully automated benchmarking suite to compare macromolecular complexes.

Protein structure prediction has a long history of benchmarking efforts such as critical assessment of structure prediction, continuous automated model evaluation and critical assessment of prediction of interactions. Read more →

From sweetener to risk factor: Network toxicology, molecular docking and molecular dynamics reveal the mechanism of aspartame in promoting coronary heart disease.

Aspartame, a widely used non-nutritive sweetener, has been epidemiologically linked to coronary heart disease (CHD), although the underlying mechanisms remain unclear. Read more →

Multi-target exploration of newly synthesized pyrazoline-quinoline derivatives via in vitro screening, QSAR, molecular docking, MD simulations, and DFT analysis.

The development of multifunctional therapeutic agents remains a promising strategy in modern drug discovery, particularly for diseases associated with oxidative stress, bacterial infections, and cancer progression. Read more →

UPLC-Q-TOF/MS-based Spectrum-effect Correlation Combined with Chemometrics and Molecular Docking for Quality Assessment and Screening of Bioactive Components with Hemostatic, Antinociceptive, and Anti-Inflammatory Activities in Liparis nervosa.

Ethnopharmacological relevance Liparis nervosa (LN) occurs in Southwest China and is traditionally used as a hemostatic and detoxifying agent; however, the pharmacodynamic basis for its medicinal properties is unclear; this impedes the quality standardization and clinical application of this herb. Read more →

Exploring the Mechanism of Platycladi Cacumen in Intervening Androgenetic Alopecia Based on Network Pharmacology, Molecular Docking, and Molecular Dynamics Simulation

Abstract As a traditional hair-growth-promoting herb, Platycladi Cacumen(PC) has a long history of folk application in the field of hair loss improvement. Read more →

A hardware demonstration of a universal programmable RRAM-based probabilistic computer for molecular docking.

Molecular docking is a critical computational strategy in drug discovery, but the diversity of biomolecular structures and flexible binding conformations create an enormous search space that challenges conventional computing. Read more →

Pipeline Tip

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

Resources & Tools

• Dataset: Protein Data Bank (PDB) - The single global archive for macromolecular structure data.
• Dataset: AlphaFold Structure Database - 200M+ predicted structures from DeepMind/EMBL-EBI.
• Tool: MultiFOLD/IntFOLD - High-performance protein structure prediction and quality assessment server. View all tools →
• Tool: PyMOL - Gold standard for molecular visualization and publication-quality imaging. View all tools →
• Event: Protein Design Hub (LinkedIn Group) (Ongoing)
• Event: Structural Biology Events (Open)
• Job: Programme Manager – Human Genomics and Translational Data (HGTD) - LinkedIn at Bioinformatics Careers
• Job: Champions Oncology, Inc. hiring Senior Director Data Engineering & Bioinformatics in United States - LinkedIn at Bioinformatics Careers

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