A gated graph transformer for protein complex structure quality assessment and its performance in CASP15.

With AlphaFold achieving high-accuracy tertiary structure prediction for most single-chain proteins (monomers), the next major challenge in protein structure prediction is accurately modeling multi-chain protein complexes (multimers). We developed MULTICOM4, the latest version of the MULTICOM system, to improve protein complex structure prediction by integrating transformer-based AlphaFold2, diffusion model-based AlphaFold3, and our in-house techniques. These include protein complex stoichiometry prediction, diverse multiple sequence alignment (MSA) generation leveraging both sequence and structure comparison, modeling exception handling, and deep learning-based model quality assessment. MULTICOM4 was blindly evaluated in the 16th community-wide Critical Assessment of Techniques for Protein Structure Prediction (CASP16) in 2024. In Phase 0 of CASP16, where stoichiometry information was unavailable, MULTICOM predictors performed best, with MULTICOM_human achieving a TM-score of 0.752 and a DockQ score of 0.584 for top-ranked predictions on average. In Phase 1 of CASP16, with stoichiometry information provided, MULTICOM_human remained among the top predictors, attaining a TM-score of 0.797 and a DockQ score of 0.558 on average. The CASP16 results demonstrate that integrating complementary AlphaFold2 and 3 with enhanced MSA inputs, comprehensive model ranking, exception handling, and accurate stoichiometry prediction can effectively improve protein complex structure prediction.

AlphaFold2 can predict protein complex structures as long as a multiple sequence alignment (MSA) of the interologs of the target protein-protein interaction (PPI) can be provided. In this study, a simplified phylogeny-based approach was applied to generate the MSA of interologs, which was then used as the input to AlphaFold2 for protein complex structure prediction. In this extensively benchmarked protocol on nonredundant PPI dataset, including 107 bacterial PPIs and 442 eukaryotic PPIs, we show complex structures of 79.5% of the bacterial PPIs and 49.8% of the eukaryotic PPIs can be successfully predicted, which yielded significantly better performance than the application of MSA of interologs prepared by two existing approaches. Considering PPIs may not be conserved in species with long evolutionary distances, we further restricted interologs in the MSA to different taxonomic ranks of the species of the target PPI in protein complex structure prediction. We found that the success rates can be increased to 87.9% for the bacterial PPIs and 56.3% for the eukaryotic PPIs if interologs in the MSA are restricted to a specific taxonomic rank of the species of each target PPI. Finally, we show that the optimal taxonomic ranks for protein complex structure prediction can be selected with the application of the predicted template modeling (TM) scores of the output models.

Protein complex structure prediction is an important problem in computational biology. While significant progress has been made for protein monomers, accurate evaluation of protein complexes remains challenging. Existing assessment methods in CASP, lack dedicated metrics for evaluating complexes. DockQ, a widely used metric, has some limitations. In this study, we propose a novel metric called BDM (Based on Distance difference Matrix) for assessing protein complex prediction structures. Our approach utilizes a distance difference matrix derived from comparing real and predicted protein structures, establishing a linear correlation with Root Mean Square Deviation (RMSD). BDM overcomes limitations associated with receptor-ligand differentiation and eliminates the requirement for structure alignment, making it a more effective and efficient metric. Evaluation of BDM using CASP14 and CASP15 test sets demonstrates superior performance compared to the official CASP scoring. BDM provides accurate and reasonable assessments of predicted protein complexes, wide adoption of BDM has the potential to advance protein complex structure prediction and facilitate related researches across scientific domains. Code is available at http://mialab.ruc.edu.cn/BDMServer/ .

SAXS Ensemble Refinement of ESCRT-III CHMP3 Conformational Transitions

Monoclonal antibodies have recently started to deliver on their promise as highly specific and active drugs; however, a more effective, knowledge-based approach to the selection, design, and optimization of potential therapeutic antibodies is currently limited by the surprising lack of detailed structural information for complexes formed with target proteins. Here we show that complexes formed with minimal antigen binding single chain variable fragments (scFv) reliably reflect all the features of the binding interface present in larger Fab fragments, which are commonly used as therapeutics, and report the development of a robust, reliable, and relatively rapid approach to the determination of high resolution models for scFv-target protein complexes. This NMR spectroscopy-based approach combines experimental determination of the interaction surfaces and relative orientations of the scFv and target protein, with NMR restraint-driven, semiflexible docking of the proteins to produce a reliable and highly informative model of the complex. Experience with scFvs and Fabs targeted at a number of secreted regulatory proteins suggests that the approach will be applicable to many therapeutic antibodies targeted at proteins, and its application is illustrated for a potential therapeutic antibody targeted at the cytokine IL-1beta. The detailed structural information that can be obtained by this approach has the potential to have a major impact on the rational design and development of an increasingly important class of biological pharmaceuticals.

Native top-down mass spectrometry (nTDMS) has emerged as a powerful structural biology tool that can localize post-translational modifications (PTMs), explore ligand-binding interactions, and elucidate the three-dimensional structure of proteins and protein complexes in the gas-phase. Fourier-transform ion cyclotron resonance (FTICR) MS offers distinct capabilities for nTDMS, owing to its ultrahigh resolving power, mass accuracy, and robust fragmentation techniques. Previous nTDMS studies using FTICR have mainly been applied to overexpressed recombinant proteins and protein complexes. Here, we report the first nTDMS study that directly analyzes human heart tissue lysate by direct infusion FTICR MS without prior chromatographic separation strategies. We have achieved comprehensive nTDMS characterization of cardiac contractile proteins that play critical roles in heart contraction and relaxation. Specifically, our results reveal structural insights into ventricular myosin light chain 2 (MLC-2v), ventricular myosin light chain 1 (MLC-1v), and alpha-tropomyosin (α-Tpm) in the sarcomere, the basic contractile unit of cardiac muscle. Furthermore, we verified the calcium (Ca2+) binding domain in MLC-2v. In summary, our nTDMS platform extends the application of FTICR MS to directly characterize the structure, PTMs, and metal-binding of endogenous proteins from heart tissue lysate without prior separation methods.

Protein-Peptide Complex Prediction through Fragment Interaction Patterns

Gas-phase protein assemblies: Unfolding landscapes and preserving native-like structures using noncovalent adducts

In recent years, the end-to-end deep learning method for single-chain protein structure prediction has achieved high accuracy. For example, the state-of-the-art method AlphaFold, developed by Google, has largely increased the accuracy of protein structure predictions to near experimental accuracy in some of the cases. At the same time, there are few methods that can evaluate the quality of protein complexes at the residue level. In particular, evaluating the quality of residues at the interface of protein complexes can lead to a wide range of applications, such as protein function analysis and drug design. In this paper, we introduce a new deep graph neural network-based method ComplexQA, to evaluate the local quality of interfaces for protein complexes by utilizing the residue-level structural information in 3D space and the sequence-level constraints. We benchmark our method to other state-of-the-art quality assessment approaches on the HAF2 and DBM55-AF2 datasets (high-quality structural models predicted by AlphaFold-Multimer), and the BM5 docking dataset. The experimental results show that our proposed method achieves better or similar performance compared with other state-of-the-art methods, especially on difficult targets which only contain a few acceptable models. Our method is able to suggest a score for each interfac e residue, which demonstrates a powerful assessment tool for the ever-increasing number of protein complexes. https://github.com/Cao-Labs/ComplexQA.git. Contact: caora@plu.edu.

Electron crystallography covers the analysis of crystals by electron microscopy based methods in order to infer into the structure of the crystallized molecules. Electron crystallography of two-dimensional crystals has become a powerful alternative to X-ray crystallography and NMR spectroscopy for the determination of the structure of proteins or protein complexes. The electron crystallography approach is particularly suitable for the study of membrane proteins, which due to their amphiphilic properties are stable in lipid bilayers, their natural environment, where they can form crystalline arrays. This method of structural analysis has also proved suitable for the study of soluble proteins, which can be induced to form two-dimensional lattices, as well as the study of very thin three-dimensional crystals. The preparation of two-dimensional crystals normally requires less protein and at a lower concentration than required for three-dimensional crystallization. Overall, the structural analysis based on electron microscopy is more amenable to the study of conformational variations associated with different functional states of membrane proteins than other techniques used in structural biology. Furthermore, the analysis of two-dimensional crystals enables the calculation of preliminary structural maps at relatively low resolutions, which at early stages in the optimization crystallization and image analysis may render valuable information on the overall protein architecture. Here we present a summary of the methods involved in protein electron crystallography, regarding both the crystallization process itself and the fundaments of electron microscopy and image analysis required. We also show how electron crystallography has greatly contributed for the development of the present knowledge on the structural organization of photosynthetic protein complexes, focusing on the studies involving transmembrane pigmented proteins and protein complexes.

Quality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. We develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method-AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method-AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA. The source code is available at https://github.com/BioinfoMachineLearning/EnQA. Supplementary data are available at Bioinformatics online.

SummaryCoevolutionary sequence analysis has become a commonly used technique for de novo prediction of the structure and function of proteins, RNA, and protein complexes. We present the EVcouplings framework, a fully integrated open-source application and Python package for coevolutionary analysis. The framework enables generation of sequence alignments, calculation and evaluation of evolutionary couplings (ECs), and de novo prediction of structure and mutation effects. The combination of an easy to use, flexible command line interface and an underlying modular Python package makes the full power of coevolutionary analyses available to entry-level and advanced users.Availability and implementationhttps://github.com/debbiemarkslab/evcouplings

The development of AlphaFold for protein structure prediction has opened a new era in structural biology. This is even more the case for AlphaFold-Multimer for the prediction of protein complexes. The interpretation of these predictions has become more important than ever, but it is difficult for the non-specialist. While an evaluation of the prediction quality is provided for monomeric protein predictions by the AlphaFold Protein Structure Database, such a tool is missing for predicted complex structures. Here, we present the PAE Viewer webserver (http://www.subtiwiki.uni-goettingen.de/v4/paeViewerDemo), an online tool for the integrated visualization of predicted protein complexes using a 3D structure display combined with an interactive representation of the Predicted Aligned Error (PAE). This metric allows an estimation of the quality of the prediction. Importantly, our webserver also allows the integration of experimental cross-linking data which helps to interpret the reliability of the structure predictions. With the PAE Viewer, the user obtains a unique online tool which for the first time allows the intuitive evaluation of the PAE for protein complex structure predictions with integrated crosslinks.

Chemical cross-linking combined with mass spectrometry is a viable approach to study the low-resolution structure of protein and protein complexes. However, unambiguous identification of the residues involved in a cross-link remains analytically challenging. To enable a more effective analysis across various MS platforms, we have developed a novel set of collision-induced dissociative cross-linking reagents and methodology for chemical cross-linking experiments using tandem mass spectrometry (CID-CXL-MS/MS). These reagents incorporate a single gas-phase cleavable bond within their linker region that can be selectively fragmented within the in-source region of the mass spectrometer, enabling independent MS/MS analysis for each peptide. Initial design concepts were characterized using a synthesized cross-linked peptide complex. Following verification and subsequent optimization of cross-linked peptide complex dissociation, our reagents were applied to homodimeric glutathione S-transferase and monomeric bovine serum albumin. Cross-linked residues identified by our CID-CXL-MS/MS method were in agreement with published crystal structures and previous cross-linking studies using conventional approaches. Common LC/MS/MS acquisition approaches such as data-dependent acquisition experiments using ion trap mass spectrometers and product ion spectral analysis using SEQUEST were shown to be compatible with our CID-CXL-MS/MS reagents, obviating the requirement for high resolution and high mass accuracy measurements to identify both intra- and interpeptide cross-links.

Molecular forces driving protein complexation of lentil and whey proteins: Structure-function relationships of trehalose-conjugated protein complexes on protein digestibility and solubility