Recep Adiyaman
Vision About Experience Education Skills Publications Tools Newsletter Updates Podcasts Blog Contact CV Subscribe
All Issues
Older Newer
Daily Signal April 23, 2026 · 9 min read

Issue #93: NNDock2: A neural network-based scoring function for ranking protein-protein docking models.

Protein Design Digest #93: NNDock2: A neural network-based scoring function for ranking protein-pro…

Share X LinkedIn
Protein Design Daily

Building something in Protein Design?

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

LinkedIn Email

Subscribe to Protein Design Digest

Daily curated signals from arXiv, PubMed, and BioRxiv.

Signal of the Day

NNDock2: A neural network-based scoring function for ranking protein-protein docking models.

Protein-protein interactions (PPIs) play crucial roles in diverse cellular functions and biological processes, and structural knowledge of the protein complexes is valuable for the elucidation of those functions and designing new drugs. Due to the limitations of experimental methods, computational modeling approaches capable of producing reliable protein complex models using molecular docking tools are of considerable practical interest. The success of protein docking largely depends on an accurate scoring function that can pick out good protein docking models. In this work, we present a neural network-based scoring function for scoring protein-protein docking models, NNDock2, the updated version of our previous scoring function, NNDock1. To improve NNDock1, we augmented the training decoys by adding a large number of more distant decoys. In addition, instead of interface root mean square deviation (iRMSD) in NNDock1, we used the fraction of native contact ([Formula: see text] as a target function, which shows better correlation with true model quality. We also applied regularization during training to avoid overfitting. We tested NNDock2 on the protein-protein docking benchmark version 5.0 (BM5), DOCKGROUND dataset, and the CAPRI score set and compared the performance of NNDock2 with other state-of-the-art scoring functions. NNDock2 performed comparably to other state-of-the-art scoring functions, despite the simplicity of the method and low computational costs. We envision that NNDock2 could be used as an independent scoring function or as an element or feature of composite or deep learning-based scoring functions for protein complex model quality estimation.

Why this matters: Expands the searchable sequence space for novel folds and high-affinity binders.

Also Worth Reading

Comprehensive Molecular Docking and Molecular Dynamics Reveal Inhibitors of HER2 L755S, T798I, and T798M based on a Large Database of Curcumin Derivatives.

Objective This study presents a methodology employing virtual screening to identify curcumin derivatives with selective affinity for the HER2 mutations L755S, T798I, and T798M. Methods Curcumin derivatives were retrieved from the ChEMBL database and filtered using KNIME. HER2 mutations were modeled in silico using MOE software with PDB ID 3RCD. Molecular docking and dynamics simulations were conducted to screen high-affinity compounds and evaluate binding interactions. Result From 505 curcumin derivatives, the RDKit module implemented in KNIME successfully filtered 317 compounds. Subsequent molecular docking against wild-type HER2 identified 100 curcumin derivatives with low docking scores, among which the top 20 compounds exhibited better binding affinities than Lapatinib. Further molecular docking screening against the three HER2 mutations identified five lead compounds with the lowest docking scores. Molecular docking and molecular dynamics simulation revealed critical binding interactions with residues essential for kinase domain stability. Chemical structural analysis revealed key modifications, such as geranyl and tripeptide modifications. CHEMBL3758656 and CHEMBL3827366, two curcumin derivatives, demonstrated consistent binding across HER2 mutations and a favorable ADMET profile. Conclusion This study successfully identified CHEMBL3758656 and CHEMBL3827366 as promising HER2 inhibitors through comprehensive virtual screening. Their high binding affinity against L755S, T798I, and T798M mutations and favorable ADME and toxicity properties underscore their potential as alternative therapeutics for HER2-positive breast cancer.

Exploring quantum frontiers in protein structure prediction: techniques, challenges, and opportunities.

Protein folding is governed by the principle of free energy minimization, where a protein’s native tertiary structure corresponds to the global minimum on an energy landscape shaped by quantum mechanical interactions such as hydrogen bonding, van der Waals forces, and electron delocalization. Despite significant advances in template-based modeling (TBM), ab-initio simulations, and deep learning approaches, classical methods continue to face challenges due to the exponential complexity of the conformational search space and the approximations involved in modeling molecular interactions. Although AlphaFold, a deep learning-based protein modeling tool, has achieved a remarkable score of 92.4 in the critical assessment of protein structure prediction (CASP), classical protein structure prediction (PSP) remains hindered by the computational limitations of conventional binary architecture in representing the physical constraints of biomolecular systems. By representing the combinatorial explosion of possible conformations as a more tractable optimization problem, quantum computing offers a fundamentally new paradigm for protein three-dimensional (3D) structure prediction. In this review, we explore how quantum computing (QC) techniques including quantum annealing, quantum optimization algorithms, and hybrid quantum-classical approaches can leverage quantum properties such as superposition, entanglement, and tunneling to more efficiently navigate the complex energy landscapes associated with protein folding. While current challenges, including limited qubit fidelity, error correction, and scalability, remain, the integration of quantum algorithms with classical strategies holds significant promise for advancing structural biology, with profound implications for drug discovery and the understanding of complex biomolecular systems.

Using experimental results of protein design to guide biomolecular energy-function development.

Computational models of macromolecules have many applications in biochemistry, but physical inaccuracies limit their utility. One class of models uses energy functions rooted in classical mechanics. The standard datasets used to train these models are limited in diversity, pointing to a need for new training data. Here, we sought to explore a new paradigm for training an energy function, where the Rosetta energy function was used to design de novo proteins. Experimental results on these designs were then used to identify failure modes of design, which were subsequently used as a “guiding principle” to retrain the energy function. Specifically, we examined a diverse set of de novo protein designs experimentally tested for their ability to stably fold, identifying unstable designs that were predicted to be stable by the Rosetta energy function. Using deep mutational scanning, we identified single amino-acid mutations that rescued the stability of these designs, providing insight into common failure modes of the energy function. We identified one key failure mode, involving steric clashing in protein cores. We identified similar overpacking when using Rosetta to refine high-resolution protein crystal structures, quantified the degree of overpacking, and refit a small set of energy-function parameters to better recapitulate native-like packing. Following fitting, we largely eliminated the failure mode in the refinement task, while retaining performance on other benchmarks, resulting in an updated version of the Rosetta energy function. This work shows how learning from protein designs can guide energy-function development.

Research & AI Updates

From the Industry

Quick Reads

Exploring the Mechanism of Oral Cancer With Shikonin Based on the Network Pharmacology and Molecular Docking Technology.

Objectives To explore the underlying mechanisms of shikonin in treating oral cancer using network pharmacology and molecular docking methods. Read more →

Unraveling the molecular interaction of hydroxytyrosol with human serum albumin via multi-spectroscopy, thermodynamic analysis, molecular docking and molecular dynamics simulation.

Hydroxytyrosol (HT) is a powerful antioxidant that scavenges free radicals and protects cells and also is one of the main active ingredients in functional foods. Read more →

Integrated Network Pharmacology, Molecular Docking, and Experimental Validation Elucidating the Therapeutic Mechanism of Idesia polycarpa Crude Oil in Aluminum Chloride-Induced Alzheimer’s Rat Models.

While mounting evidence points to a potential link between industrial aluminum exposure and neurodegenerative diseases like Alzheimer’s disease (AD), the precise intervention strategies remain an area of active research. Read more →

Network toxicology and molecular docking elucidation of oligoasthenospermia induced by bisphenol analogues: a comparative study of BPA, BPS, BPF, and BPAF.

Bisphenol A (BPA) substitutes are increasingly prevalent in consumer products, yet their potential to induce oligoasthenospermia (OAS) remains poorly understood. Read more →

Plants play a vital role in sustaining life on Earth, offering essential resources such as food, energy, shelter, and serving as a key source for drug development. Read more →

Network Pharmacology, Molecular Docking, and Experimental Verification Employed to Elucidate the Potential Mechanisms of Bazi Bushen Capsule for the Treatment of Premature Ovarian Failure.

Introduction The aim of this study was to investigate the mechanism of BZBS in the treatment of POF using network pharmacology and molecular docking, and to validate it through in vivo experiments. Read more →

Harnessing machine learning and multi-scale modeling to discover novel ALOX15 inhibitors from marine natural products.

ALOX15 is a key regulatory enzyme in multiple pathological processes including inflammation, cancer, and cardiovascular disease, rendering the development of potent inhibitors of this enzyme of significant clinical importance. Read more →

Comprehensive Exploration of Three Pistachio Varieties: Antidiabetic, Anticholinergic, Antioxidant Properties, Phytochemical Profile by LC-MS/MS, and Molecular Docking.

Pistachios (Pistacia vera L.) are a significant food that is added to many foods and also consumed as a snack. Read more →

Pipeline Tip

Employ HADDOCK for ambiguous restraints in protein-protein docking.

Resources & Tools

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

BS HF DK