Aug 2025· Journal of Chemical Information and Modeling· Vol 65, pp. 8497-8513· 13 citations· ⚡ 1 influential· 60 references
Computer ScienceMedicine
TL;DR
PepPCBench enables a robust evaluation of PFNN-based methods and supports their continued development for peptide-protein structure prediction, and highlights the influence of peptide length, conformational flexibility, and training set similarity on prediction accuracy.
Abstract
Accurate modeling of protein-peptide interactions is essential for understanding fundamental biological processes and designing peptide-based drugs. However, predicting the complex structures of these interactions remains challenging, primarily due to the high conformational flexibility of peptides. To support a fair and systematic evaluation of recent deep learning (DL) approaches, we introduce PepPCBench, a benchmarking framework tailored to assess protein folding neural networks (PFNNs) in protein-peptide complex prediction. As part of this framework, we curated PepPCSet, a data set of 261 experimentally resolved complexes with peptides ranging from 5 to 30 residues. We benchmark five full-atom PFNNs, including AlphaFold3 (AF3), AlphaFold-Multimer (AFM), Chai-1, HelixFold3 (HF3), and RoseTTAFold-All-Atom (RFAA), using comprehensive evaluation metrics. Our benchmarking reveals meaningful performance differences among these methods and highlights the influence of peptide length, conformational flexibility, and training set similarity on prediction accuracy. While AF3 shows strong performance in structure prediction, further analysis indicates that confidence metrics correlate poorly with experimental binding affinities, underscoring the need for improved scoring strategies and generalizability. By providing a reproducible and extensible framework, PepPCBench enables a robust evaluation of PFNN-based methods and supports their continued development for peptide-protein structure prediction.
P PepXPro is presented, a modular framework that transforms publicly available protein-peptide structure-affinity resources into curated datasets and reproducible benchmark collections generated under user-defined criteria that provides an extensible foundation for reproducible protein- peptide benchmark construction.
Deep learning-based protein structure prediction methods that leverage evolutionary information from multiple sequence alignments (MSAs), exemplified by AlphaFold2, have achieved remarkable accuracy. However, existing methods still struggle to predict challenging proteins, particularly those with novel folds or limited evolutionary information, and to recover alternative conformational states. Here we show that structure prediction models trained under different MSA-depth distributions corresponding to different levels of evolutionary information exhibit complementary generalization behaviors, and that a model trained on a mixture of these distributions can combine their complementary generalization strengths. Building on this insight, we developed ProtMonomer, a deep learning framework trained on MSA-depth distributions representing a broad range of evolutionary information levels to improve structure prediction. Across benchmarks comprising CASP15 targets, non-redundant experimentally determined structures, orphan proteins, and short peptides, ProtMonomer performed comparably to or better than leading methods, including AlphaFold2 and AlphaFold3, with particularly strong performance on challenging targets. For fold-switching proteins, ProtMonomer also recovered alternative conformational states more accurately than AlphaFold2 and AlphaFold3 across diverse homologous sequence sampling strategies. In addition to improving predictive accuracy, ProtMonomer substantially reduced inference cost through an efficient architecture, enabling high-throughput applications. Together, these findings provide insights into the generalization of evolution-informed structure prediction models and support ProtMonomer as an accurate and efficient framework for protein structure prediction.
Abstract Predicting accurate protein structures is essential for understanding molecular mechanisms, interpreting the impact of sequence variation, and supporting translational applications ranging from drug discovery to clinical genomics. Recent advances in deep-learning–based predictors such as AlphaFold2, OpenFold, and AlphaFold3 have transformed structural biology, enabling routine in silico modeling even for challenging or previously uncharacterized proteins. However, systematic benchmarking of these tools—especially for novel targets and single amino acid variants—remains limited. Conventional global metrics often fail to capture biologically meaningful discrepancies. By evaluating multiple implementations of AlphaFold2 and OpenFold, together with ColabFold and the AlphaFold3 server, across 10 different proteins and 222 single amino acid protein variants encompassing a wide range of sizes, structures, and functions, we show that although widely used global indicators—like mean pLDDT, pTM-score, and RMSD—frequently suggest comparable performance, substantial local-level differences remain elusive. To address this gap, we introduce a comparative framework leveraging Bland–Altman agreement analysis, to evaluate per-residue Cα-confidence differences and Per-Residue profiles (PRPs), complemented by Uniform Manifold Approximation and Projection (UMAP). This approach reveals marked localized divergences, particularly within flexible or intrinsically disordered regions, where both predictor choice and single-residue substitutions trigger the largest conformational shifts. We further demonstrate that using reduced homology databases has minimal impact on predicted structural quality, offering computationally efficient alternatives. Collectively, our findings underscore the importance of integrating global and residue-specific evaluations to more accurately assess robustness, agreement, and practical usability across contemporary protein structure prediction methods.
Florencia R. Díaz, Daniela Orschanski, Juan I. Folco et al.· Briefings in Bioinformatics· 0 citations
Findings identify pose ranking, rather than pose generation, as the major limitation of current cyclic peptide–protein complex prediction and demonstrate that complementary structural features can improve confidence-based pose selection.
Zhe Li, Ye Yuan, Kaiqiang Hu et al.· bioRxiv· 0 citations
Deep learning methods, such as AlphaFold and RosettaFold, achieve high accuracy in protein structure prediction. However, predicting the structure of large protein complexes remains challenging due to their large size and intricate multi-chain interactions. Docking-based methods can handle large proteins, but are limited by the huge combinatorial binding space of multichains. Assembly-based approaches offer an alternative, but their accuracy critically relies on the precision of predicted subcomponents. Addressing the challenges, we propose HDOCK-Multimer (HDM), a structure prediction framework of large protein complexes by integrating ab initio docking and combinatorial assembly. HDM can efficiently reduce reliance on subcom-ponent accuracy through docking process, while leveraging the pairwise interactions of subcom-ponents through assembly strategy. HDM is extensively validated on three benchmarks of 35 large heteromeric complexes, 172 large protein complexes, and 7 CASP15 targets, and compared with state-of-the-art methods including MoLPC, CombFold, AlphaFold-Multimer (AFM), and AlphaFold3 (AF3). It is shown that HDOCK-Multimer substantially outperforms the other methods. In addition, HDM also shows ability to predict the stoichiometry and model the complex without stoichiometry input. It is anticipated that HDM will serve as a powerful tool for studying large protein complexes or molecular machines. The HDM package is freely available at https://github.com/huang-laboratory/HDOCK-Multimer/.
Xuan Yao, Yifan Ya, Hao Li et al.· bioRxiv· 0 citations
These findings provide practical guidance for integrating open-source protein structure prediction models into AI-driven nanobody discovery pipelines while highlighting the need for improved generalization across antigens.
Yannick Vogt, Rebekka Roßberg, Jan Habermann et al.· Frontiers in Bioinformatics· 1 citation
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduAug 18, 2026
A new method for surgically removing training examples from a model reveals that as datasets grow, the link between what a model learns and what it produces dissolves.
What does it take to trust AI-driven HVAC optimization? Our AI Model Factory combines agents, machine learning, reinforcement learning and deterministic checks in a governed workflow designed for messy, real-world building data. The post We built an AI factory for HVAC control appeared first on GPT-Lab.
MIT News · Artificial Intelligence· news.mit.eduAug 27, 2026
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.
MIT News · Artificial Intelligence· news.mit.eduAug 3, 2026
