Improved protein structure refinement guided by deep learning based accuracy estimation

Biomolecular conformational changes and transient druggable binding sites through full-length AMPK molecular dynamics simulations

AMPK (AMP-activated protein kinase) is a crucial signaling protein found in essentially all eukaryotic organisms and acts as an energy sensor. When activated by metabolic stress, AMPK phosphorylates a variety of molecular targets, altering enzyme activity and gene expression to regulate cellular responses. In general, in response to low intracellular ATP levels (high ADP:ATP ratio), AMPK triggers the activation of energy-producing pathways while simultaneously inhibiting energy-consuming processes. Recent studies have established a connection between molecular pathways involved in sensing energy and potential for extending longevity. AMPK indirect activator compounds have shown a potential strategy to obtain an anti-aging biological activity. This study explores the conformational changes and transient druggable binding pockets over the 1 μs trajectory of molecular dynamics simulations to comprehend the behavior of main domains and allosteric drug and metabolite (ADaM) site. The described conformations of the apo-ADaM site suggest an important influence of specific residues on the cavity volume variations. A clustering set of representative AMPK conformations allowed to identify the more favorable binding site volume and shape at the protein apo form, including the carbohydrate-binding module (CBM) region which exhibited a stable movement near the ADaM site of the alpha-subunit. The identification of gamma-subunit transient druggable binding pocket CBS3 during the microscale time trajectory simulations also offers valuable insights into structure-based AMP-mimetic drug design for AMPK activation.

AI-based quality assessment methods for protein structure models from cryo-EM

Cryogenic electron microscopy (cryo-EM) has revolutionized structural biology, with an increasing number of structures being determined by cryo-EM each year, many at higher resolutions. However, challenges remain in accurately interpreting cryo-EM maps. Inaccuracies can arise in regions of locally low resolution, where manual model building is more prone to errors. Validation scores for structure models have been developed to assess both the compatibility between map density and the structure, as well as the geometric and stereochemical properties of protein models. Recent advancements have introduced artificial intelligence (AI) into this field. These emerging AI-driven tools offer unique capabilities in the validation and refinement of cryo-EM-derived protein atomic models, potentially leading to more accurate protein structures and deeper insights into complex biological systems.

Next generation technologies for protein structure determination: challenges and breakthroughs in plant biology applications

Advancements in structural biology have significantly deepened our understanding of plant proteins, which are central to critical biological functions such as photosynthesis, metabolism, signal transduction, and structural architechture. Gaining insights into their structures is crucial for unraveling their functions and mechanisms, which in turn has profound implications for agriculture, biotechnology, and environmental sustainability. Traditional methods in protein structural biology often fall short in addressing large protein assemblies and membrane proteins, and, in particular the dynamics and structural features of proteins in the native cellular context. This paper explores how next-generation technologies are transforming the field of plant protein structural biology, offering powerful tools to overcome longstanding obstacles and enabling remarkable scientific breakthroughs. Key technologies discussed include advanced X-ray crystallography, Cryo-Electron microscopy, Nuclear Magnetic Resonance spectroscopy, Cross-linking mass spectrometry, and Artificial Intelligence-driven approaches. These technologies are examined in terms of their challenges, innovations, and application with particular emphasis on their relevance to plant systems. Future directions in plant protein structural biology are also discussed. Although technical details are not covered in depth, readers are referred to the primary literature for more comprehensive information.

M-DeepAssembly: enhanced DeepAssembly based on multi-objective multi-domain protein conformation sampling

BACKGROUND

Association and cooperation among structural domains play an important role in protein function and drug design. Despite remarkable advancements in highly accurate single-domain protein structure prediction through the collaborative efforts of the community using deep learning, challenges still exist in predicting multi-domain protein structures when the evolutionary signal for a given domain pair is weak or the protein structure is large.

RESULTS

To alleviate the above challenges, we proposed M-DeepAssembly, a protocol based on multi-objective protein conformation sampling algorithm for multi-domain protein structure prediction. Firstly, the inter-domain interactions and full-length sequence distance features are extracted through DeepAssembly and AlphaFold2, respectively. Secondly, subject to these features, we constructed a multi-objective energy model and designed a sampling algorithm for exploring and exploiting conformational space to generate ensembles. Finally, the output protein structure was selected from the ensembles using our in-house developed model quality assessment algorithm. On the test set of 164 multi-domain proteins, the results show that the average TM-score of M-DeepAssembly is 15.4% and 2.0% higher than AlphaFold2 and DeepAssembly, respectively. It is worth noting that there are models with higher accuracy in ensembles, achieving an improvement of 20.3% and 6.4% relative to the two baseline methods, although these models were not selected. Furthermore, when compared to the prediction results of AlphaFold2 for CASP15 multi-domain targets, M-DeepAssembly demonstrates certain performance advantages.

CONCLUSIONS

M-DeepAssembly provides a distinctive multi-domain protein assembly algorithm, which can alleviate the current challenges of weak evolutionary signals and large structures to some extent by forming diverse ensembles using multi-objective protein conformation sampling algorithm. The proposed method contributes to exploring the functions of multi-domain proteins, especially providing new insights into targets with multiple conformational states.

LightRoseTTA: High-Efficient and Accurate Protein Structure Prediction Using a Light-Weight Deep Graph Model

Accurately predicting protein structure, from sequences to 3D structures, is of great significance in biological research. To tackle this issue, a representative deep big model, RoseTTAFold, is proposed with promising success. Here, "a light-weight deep graph network, named LightRoseTTA," is reported to achieve accurate and highly efficient prediction for proteins. Notably, three highlights are possessed by LightRoseTTA: i) high-accurate structure prediction for proteins, being "competitive with RoseTTAFold" on multiple popular datasets including CASP14 and CAMEO; ii) high-efficient training and inference with a light-weight model, costing "only 1 week on one single NVIDIA 3090 GPU for model-training" (vs 30 days on 8 NVIDIA V100 GPUs for RoseTTAFold) and containing "only 1.4M parameters" (vs 130M in RoseTTAFold); iii) low dependency on multi-sequence alignment (MSA), achieving the best performance on three MSA-insufficient datasets: Orphan, De novo, and Orphan25. Besides, LightRoseTTA is "transferable" from general proteins to antibody data, as verified in the experiments. The time and resource costs of LightRoseTTA and RoseTTAFold are further discussed to demonstrate the feasibility of light-weight models for protein structure prediction, which may be crucial in resource-limited research for universities and academic institutions. The code and model are released to speed biological research (https://github.com/psp3dcg/LightRoseTTA).

$\mathcal{S}$ able: bridging the gap in protein structure understanding with an empowering and versatile pre-training paradigm

Protein pre-training has emerged as a transformative approach for solving diverse biological tasks. While many contemporary methods focus on sequence-based language models, recent findings highlight that protein sequences alone are insufficient to capture the extensive information inherent in protein structures. Recognizing the crucial role of protein structure in defining function and interactions, we introduce $\mathcal{S}$able, a versatile pre-training model designed to comprehensively understand protein structures. $\mathcal{S}$able incorporates a novel structural encoding mechanism that enhances inter-atomic information exchange and spatial awareness, combined with robust pre-training strategies and lightweight decoders optimized for specific downstream tasks. This approach enables $\mathcal{S}$able to consistently outperform existing methods in tasks such as generation, classification, and regression, demonstrating its superior capability in protein structure representation. The code and models can be accessed via GitHub repository at https://github.com/baaihealth/Sable.

Design of high-affinity binders to immune modulating receptors for cancer immunotherapy

Immune receptors have emerged as critical therapeutic targets for cancer immunotherapy. Designed protein binders can have high affinity, modularity, and stability and hence could be attractive components of protein therapeutics directed against these receptors, but traditional Rosetta based protein binder methods using small globular scaffolds have difficulty achieving high affinity on convex targets. Here we describe the development of helical concave scaffolds tailored to the convex target sites typically involved in immune receptor interactions. We employed these scaffolds to design proteins that bind to TGFβRII, CTLA-4, and PD-L1, achieving low nanomolar to picomolar affinities and potent biological activity following experimental optimization. Co-crystal structures of the TGFβRII and CTLA-4 binders in complex with their respective receptors closely match the design models. These designs should have considerable utility for downstream therapeutic applications.

Protein-surface interactions in nano-scale biosensors for IL-6 detection using functional monolayers

A multiscale approach is employed to investigate the interaction dynamics between interleukin-6, a key cancer biomarker, and alkyl-functionalized surfaces, with the ultimate goal of guiding biosensor design. The study integrates classical molecular dynamics, Brownian dynamics simulations, and binding experiments to explore the adsorption dynamics and energetics of IL-6 on surfaces modified with self-assembled monolayers (SAMs). The comparative analysis reveals a dramatic effect on the interaction strength of IL-6 with a SAMs comprising a mix of charged and hydrophobic ligands. Solvent accessible surface area analysis shows enhanced exposure of charged terminal groups on the mixed SAM surface. Experimental investigations using surface plasmon resonance reveal that IL-6 interactions enhance with increased charged ligand content in mixed SAMs, retaining high binding affinity even under high ionic strength conditions. Computational studies further highlight hydrophobic and electrostatic interactions as key factors driving the high affinity of IL-6 on the mixed SAMs surface. This research offers insights into optimizing surfaces for enhanced IL-6 recognition, which can be extended to other protein biomarkers, by combining experimental and computational approaches to improve biosensing performance.

Peripheral membrane protein endophilin B1 probes, perturbs and permeabilizes lipid bilayers

Bin/Amphiphysin/Rvs167 (BAR) domain containing proteins are peripheral membrane proteins that regulate intracellular membrane curvature. BAR protein endophilin B1 plays a key role in multiple cellular processes critical for oncogenesis, including autophagy and apoptosis. Amphipathic regions in endophilin B1 drive membrane association and tubulation through membrane scaffolding. Our understanding of exactly how BAR proteins like endophilin B1 promote highly diverse intracellular membrane remodeling events in the cell is severely limited due to lack of high-resolution structural information. Here we present the highest resolution cryo-EM structure of a BAR protein to date and the first structures of a BAR protein bound to a lipid bicelle. Using neural networks, we can effectively sort particle species of different stoichiometries, revealing the tremendous flexibility of post-membrane binding, pre-polymer BAR dimer organization and membrane deformation. We also show that endophilin B1 efficiently permeabilizes negatively charged liposomes that contain mitochondria-specific lipid cardiolipin and propose a new model for Bax-mediated cell death.

Physical-aware model accuracy estimation for protein complex using deep learning method

With the breakthrough of AlphaFold2 on monomers, the research focus of structure prediction has shifted to protein complexes, driving the continued development of new methods for multimer structure prediction. Therefore, it is crucial to accurately estimate quality scores for the multimer model independent of the used prediction methods. In this work, we propose a physical-aware deep learning method, DeepUMQA-PA, to evaluate the residue-wise quality of protein complex models. Given the input protein complex model, the residue-based contact area and orientation features were first constructed using Voronoi tessellation, representing the potential physical interactions and hydrophobic properties. Then, the relationship between local residues and the overall complex topology as well as the inter-residue evolutionary information are characterized by geometry-based features, protein language model embedding representation, and knowledge-based statistical potential features. Finally, these features are fed into a fused network architecture employing equivalent graph neural network and ResNet network to estimate residue-wise model accuracy. Experimental results on the CASP15 test set demonstrate that our method outperforms the state-of-the-art method DeepUMQA3 by 3.69 % and 3.49 % on Pearson and Spearman, respectively. Notably, our method achieved 16.8 % and 15.5 % improvement in Pearson and Spearman, respectively, for the evaluation of nanobody-antigens. In addition, DeepUMQA-PA achieved better MAE scores than AlphaFold-Multimer and AlphaFold3 self-assessment methods on 43 % and 50 % of the targets, respectively. All these results suggest that physical-aware information based on the area and orientation of atom-atom and atom-solvent contacts has the potential to capture sequence-structure-quality relationships of proteins, especially in the case of flexible proteins. The DeepUMQA-PA server is freely available at http://zhanglab-bioinf.com/DeepUMQA-PA/.

AlphaFold-guided molecular replacement for solving challenging crystal structures

Molecular replacement (MR) is highly effective for biomolecular crystal structure determination, increasingly so as the database of known structures has increased. For candidates without recognizable similarity to known structures, however, crystal structure analyses have nearly always required experiments for de novo phase evaluation. Now, with the unprecedented accuracy of AlphaFold predictions of protein structures from amino-acid sequences, an appreciable expansion of the reach of MR for proteins is realized. Here, we sought to automate an AlphaFold-guided MR procedure that tailors predictions to the MR problem at hand. We first optimized the reliability cutoff parameters for residue inclusion as tested in application to a previously MR-intractable problem. We then examined cases where AlphaFold by default predicts a conformation alternative to that of the candidate structure, devising tests for MR solution either from domain-specific predictions or from predictions based on diverse sequence subclusters. We tested subclustering procedures on an enzyme system that entails multiple MR-challenging conformations. The overall process as implemented in Phenix automatically surveys a succession of trials of increasing computational complexity until an MR solution is found or the options are exhausted. Validated MR solutions were found for 92% of one set of 158 challenging problems from the PDB and 93% of those from a second set of 215 challenges. Thus, many crystal structure analyses that previously required experimental phase evaluation can now be solved by AlphaFold-guided MR. In effect, this and related MR approaches are de novo phasing methods.

Design of pseudosymmetric protein hetero-oligomers

Pseudosymmetric hetero-oligomers with three or more unique subunits with overall structural (but not sequence) symmetry play key roles in biology, and systematic approaches for generating such proteins de novo would provide new routes to controlling cell signaling and designing complex protein materials. However, the de novo design of protein hetero-oligomers with three or more distinct chains with nearly identical structures is a challenging unsolved problem because it requires the accurate design of multiple protein-protein interfaces simultaneously. Here, we describe a divide-and-conquer approach that breaks the multiple-interface design challenge into a set of more tractable symmetric single-interface redesign tasks, followed by structural recombination of the validated homo-oligomers into pseudosymmetric hetero-oligomers. Starting from de novo designed circular homo-oligomers composed of 9 or 24 tandemly repeated units, we redesigned the inter-subunit interfaces to generate 19 new homo-oligomers and structurally recombined them to make 24 new hetero-oligomers, including ABC heterotrimers, A2B2 heterotetramers, and A3B3 and A2B2C2 heterohexamers which assemble with high structural specificity. The symmetric homo-oligomers and pseudosymmetric hetero-oligomers generated for each system have identical or nearly identical backbones, and hence are ideal building blocks for generating and functionalizing larger symmetric and pseudosymmetric assemblies.

Recent advances and challenges in protein complex model accuracy estimation

Estimation of model accuracy plays a crucial role in protein structure prediction, aiming to evaluate the quality of predicted protein structure models accurately and objectively. This process is not only key to screening candidate models that are close to the real structure, but also provides guidance for further optimization of protein structures. With the significant advancements made by AlphaFold2 in monomer structure, the problem of single-domain protein structure prediction has been widely solved. Correspondingly, the importance of assessing the quality of single-domain protein models decreased, and the research focus has shifted to estimation of model accuracy of protein complexes. In this review, our goal is to provide a comprehensive overview of the reference and statistical metrics, as well as representative methods, and the current challenges within four distinct facets (Topology Global Score, Interface Total Score, Interface Residue-Wise Score, and Tertiary Residue-Wise Score) in the field of complex EMA.

Nonsense CD247 mutations show dominant-negative features in T-cell receptor expression and function

BACKGROUND

The invariant TCR ζ/CD247 homodimer is crucial for TCR/CD3 expression and signaling through its 3 immunoreceptor tyrosine-based activation motifs (ITAMs). Homozygous null mutations in CD247 lead to immunodeficiency, while carriers exhibit 50% reduced surface CD3. It is unclear whether carriers of other CD247 variants show dominant-negative effects.

OBJECTIVE

We sought to analyze and model the potential impact on T-cell receptor (TCR) expression and function of heterozygous nonsense CD247 mutations found in patients with signs of immunodeficiency or autoimmunity.

METHODS

Jurkat T cells, either wild-type (WT) or CRISPR/Cas9-edited CD247-deficient (ZKO), were lentivirally transduced with WT CD247 or mutations ablating 1 (Q142X), 2 (Q101X), or 3 (Q70X) ITAMs.

RESULTS

Three patients from unrelated families were studied. Two heterozygous nonsense CD247 mutations were identified (p.Y152X and p.Q101X), which affected ITAM-3 and ITAM-2 and ITAM-3, respectively. Both mutations were associated with low surface CD3 expression and normal intracellular CD247 levels using a transmembrane-specific antibody, but very low intracellular CD247 levels using an ITAM-3-specific one, suggesting the presence of truncated variants in T cells. Transduction of the mutations lacking 1, 2, or 3 ITAMs into ZKO cells could not restore normal surface CD3 expression (only 60%, 22%, and 10%, respectively), whereas in WT cells, normal surface CD3 expression was reduced (to 39%, 19%, and 9% of normal levels), and both effects were dependent on ITAM number. All 6 transfectants showed reduced CD69 induction (25% to 50%), indicating that they were unable to signal downstream properly, neither isolated nor associated with WT CD247.

CONCLUSIONS

Our results suggest that CD247 variants lacking ITAMs due to nonsense, but not null, mutations are defective for normal TCR assembly and exert a dominant-negative effect on TCR expression and signaling in vitro. This, in turn, may correlate with clinical features in vivo.

Prospects for global sustainable development through integrating the environmental impacts of economic activities

Human economic activities drive the production and consumption of goods and services, contribute to the achievement of the United Nations Sustainable Development Goals (SDGs). However, the extent of economic growth's influence on the SDGs remains unclear. To fill this knowledge gap, here, we quantified the environmental effects of economic activities and explored correlations between environmental effect and achieving SDGs. We developed six Environmental Footprint Indices, with a higher score indicating better efficiency or lower burden. Here we show that the various Environmental Footprint Indices had synergistic and trade-off effects on most SDG targets indices, but the synergistic effects prevailed. As income increased, the correlation between Environmental Footprint Indices and SDG target indices gradually strengthened. improved production efficiency and consumption changes notably advance SDGs, especially in low-income group countries. Our work provides scientific insights into the impact and prospects of environmental regulation required for achieving the SDGs by 2030.

Modeling protein-small molecule conformational ensembles with ChemNet

Modeling the conformational heterogeneity of protein-small molecule systems is an outstanding challenge. We reasoned that while residue level descriptions of biomolecules are efficient for de novo structure prediction, for probing heterogeneity of interactions with small molecules in the folded state an entirely atomic level description could have advantages in speed and generality. We developed a graph neural network called ChemNet trained to recapitulate correct atomic positions from partially corrupted input structures from the Cambridge Structural Database and the Protein Data Bank; the nodes of the graph are the atoms in the system. ChemNet accurately generates structures of diverse organic small molecules given knowledge of their atom composition and bonding, and given a description of the larger protein context, and builds up structures of small molecules and protein side chains for protein-small molecule docking. Because ChemNet is rapid and stochastic, ensembles of predictions can be readily generated to map conformational heterogeneity. In enzyme design efforts described here and elsewhere, we find that using ChemNet to assess the accuracy and pre-organization of the designed active sites results in higher success rates and higher activities; we obtain a preorganized retroaldolase with a k cat/K M of 11000 M-1min-1, considerably higher than any pre-deep learning design for this reaction. We anticipate that ChemNet will be widely useful for rapidly generating conformational ensembles of small molecule and small molecule-protein systems, and for designing higher activity preorganized enzymes.

All-Atom Protein Sequence Design Based on Geometric Deep Learning

Designing sequences for specific protein backbones is a key step in creating new functional proteins. Here, we introduce GeoSeqBuilder, a deep learning framework that integrates protein sequence generation with side chain conformation prediction to produce the complete all-atom structures for designed sequences. GeoSeqBuilder uses spatial geometric features from protein backbones and explicitly includes three-body interactions of neighboring residues. GeoSeqBuilder achieves native residue type recovery rate of 51.6 %, comparable to ProteinMPNN and other leading methods, while accurately predicting side chain conformations. We first used GeoSeqBuilder to design sequences for thioredoxin and a hallucinated three-helical bundle protein. All the 15 tested sequences expressed as soluble monomeric proteins with high thermal stability, and the 2 high-resolution crystal structures solved closely match the designed models. The generated protein sequences exhibit low similarity (minimum 23 %) to the original sequences, with significantly altered hydrophobic cores. We further redesigned the hydrophobic core of glutathione peroxidase 4, and 3 of the 5 designs showed improved enzyme activity. Although further testing is needed, the high experimental success rate in our testing demonstrates that GeoSeqBuilder is a powerful tool for designing novel sequences for predefined protein structures with atomic details. GeoSeqBuilder is available at https://github.com/PKUliujl/GeoSeqBuilder.

ModFOLD9: A Web Server for Independent Estimates of 3D Protein Model Quality

Accurate models of protein tertiary structures are now available from numerous advanced prediction methods, although the accuracy of each method often varies depending on the specific protein target. Additionally, many models may still contain significant local errors. Therefore, reliable, independent model quality estimates are essential both for identifying errors and selecting the very best models for further biological investigations. ModFOLD9 is a leading independent server for detecting the local errors in models produced by any method, and it can accurately discriminate between high-quality models from multiple alternative approaches. ModFOLD9 incorporates several new scores from deep learning-based approaches, leading to greatly improved prediction accuracy compared with earlier versions of the server. ModFOLD9 is continuously independently benchmarked, and it is shown to be highly competitive with other public servers. ModFOLD9 is freely available at https://www.reading.ac.uk/bioinf/ModFOLD/.

AlphaFold two years on: Validation and impact

Two years on from the initial release of AlphaFold, we have seen its widespread adoption as a structure prediction tool. Here, we discuss some of the latest work based on AlphaFold, with a particular focus on its use within the structural biology community. This encompasses use cases like speeding up structure determination itself, enabling new computational studies, and building new tools and workflows. We also look at the ongoing validation of AlphaFold, as its predictions continue to be compared against large numbers of experimental structures to further delineate the model's capabilities and limitations.

Artificial intelligence in fusion protein three-dimensional structure prediction: Review and perspective

Recent advancements in artificial intelligence (AI) have accelerated the prediction of unknown protein structures. However, accurately predicting the three-dimensional (3D) structures of fusion proteins remains a difficult task because the current AI-based protein structure predictions are focused on the WT proteins rather than on the newly fused proteins in nature. Following the central dogma of biology, fusion proteins are translated from fusion transcripts, which are made by transcribing the fusion genes between two different loci through the chromosomal rearrangements in cancer. Accurately predicting the 3D structures of fusion proteins is important for understanding the functional roles and mechanisms of action of new chimeric proteins. However, predicting their 3D structure using a template-based model is challenging because known template structures are often unavailable in databases. Deep learning (DL) models that utilize multi-level protein information have revolutionized the prediction of protein 3D structures. In this review paper, we highlighted the latest advancements and ongoing challenges in predicting the 3D structure of fusion proteins using DL models. We aim to explore both the advantages and challenges of employing AlphaFold2, RoseTTAFold, tr-Rosetta and D-I-TASSER for modelling the 3D structures. HIGHLIGHTS: This review provides the overall pipeline and landscape of the prediction of the 3D structure of fusion protein. This review provides the factors that should be considered in predicting the 3D structures of fusion proteins using AI approaches in each step. This review highlights the latest advancements and ongoing challenges in predicting the 3D structure of fusion proteins using deep learning models. This review explores the advantages and challenges of employing AlphaFold2, RoseTTAFold, tr-Rosetta, and D-I-TASSER to model 3D structures.

AI-Driven Deep Learning Techniques in Protein Structure Prediction

Protein structure prediction is important for understanding their function and behavior. This review study presents a comprehensive review of the computational models used in predicting protein structure. It covers the progression from established protein modeling to state-of-the-art artificial intelligence (AI) frameworks. The paper will start with a brief introduction to protein structures, protein modeling, and AI. The section on established protein modeling will discuss homology modeling, ab initio modeling, and threading. The next section is deep learning-based models. It introduces some state-of-the-art AI models, such as AlphaFold (AlphaFold, AlphaFold2, AlphaFold3), RoseTTAFold, ProteinBERT, etc. This section also discusses how AI techniques have been integrated into established frameworks like Swiss-Model, Rosetta, and I-TASSER. The model performance is compared using the rankings of CASP14 (Critical Assessment of Structure Prediction) and CASP15. CASP16 is ongoing, and its results are not included in this review. Continuous Automated Model EvaluatiOn (CAMEO) complements the biennial CASP experiment. Template modeling score (TM-score), global distance test total score (GDT_TS), and Local Distance Difference Test (lDDT) score are discussed too. This paper then acknowledges the ongoing difficulties in predicting protein structure and emphasizes the necessity of additional searches like dynamic protein behavior, conformational changes, and protein-protein interactions. In the application section, this paper introduces some applications in various fields like drug design, industry, education, and novel protein development. In summary, this paper provides a comprehensive overview of the latest advancements in established protein modeling and deep learning-based models for protein structure predictions. It emphasizes the significant advancements achieved by AI and identifies potential areas for further investigation.

Protein Retrieval via Integrative Molecular Ensembles (PRIME) through Extended Similarity Indices

Molecular dynamics (MD) simulations are ideally suited to describe conformational ensembles of biomolecules such as proteins and nucleic acids. Microsecond-long simulations are now routine, facilitated by the emergence of graphical processing units. Clustering, which groups objects based on structural similarity, is typically used to process ensembles, leading to different states, their populations, and the identification of representative structures. A popular pipeline combines hierarchical clustering for clustering and selecting the cluster centroid as representative of the cluster. Here, we propose to improve on this approach, by developing a module-Protein Retrieval via Integrative Molecular Ensembles (PRIME), that consists of tools to improve the prediction of the representative in the most populated cluster using extended continuous similarity. PRIME is integrated with our Molecular Dynamics Analysis with N-ary Clustering Ensembles (MDANCE) package and can be used as a postprocessing tool for arbitrary clustering algorithms, compatible with several MD suites. PRIME predictions produced structures that when aligned to the experimental structure were better superposed (lower RMSD). A further benefit of PRIME is its linear scaling─rather than the traditional O(N2) traditionally associated with comparisons of elements in a set.

De Novo Design of Integrin α5β1 Modulating Proteins for Regenerative Medicine

Integrin α5β1 is crucial for cell attachment and migration in development and tissue regeneration, and α5β1 binding proteins could have considerable utility in regenerative medicine and next-generation therapeutics. We use computational protein design to create de novo α5β1-specific modulating miniprotein binders, called NeoNectins, that bind to and stabilize the open state of α5β1. When immobilized onto titanium surfaces and throughout 3D hydrogels, the NeoNectins outperform native fibronectin and RGD peptide in enhancing cell attachment and spreading, and NeoNectin-grafted titanium implants outperformed fibronectin and RGD-grafted implants in animal models in promoting tissue integration and bone growth. NeoNectins should be broadly applicable for tissue engineering and biomedicine.

Sequence, structure prediction, and epitope analysis of the polymorphic membrane protein family in Chlamydia trachomatis

The polymorphic membrane proteins (Pmps) are a family of autotransporters that play an important role in infection, adhesion and immunity in Chlamydia trachomatis. Here we show that the characteristic GGA(I,L,V) and FxxN tetrapeptide repeats fit into a larger repeat sequence, which correspond to the coils of a large beta-helical domain in high quality structure predictions. Analysis of the protein using structure prediction algorithms provided novel insight to the chlamydial Pmp family of proteins. While the tetrapeptide motifs themselves are predicted to play a structural role in folding and close stacking of the beta-helical backbone of the passenger domain, we found many of the interesting features of Pmps are localized to the side loops jutting out from the beta helix including protease cleavage, host cell adhesion, and B-cell epitopes; while T-cell epitopes are predominantly found in the beta-helix itself. This analysis more accurately defines the Pmp family of Chlamydia and may better inform rational vaccine design and functional studies.

Atom-ProteinQA: Atom-level protein model quality assessment through fine-grained joint learning

MOTIVATION

Protein model quality assessment (ProteinQA) is a fundamental task that is essential for biologically relevant applications, i.e., protein structure refinement, protein design, etc. Previous works aimed to conduct ProteinQA only on the global structure or per-residue level, ignoring potentially usable and precise cues from a fine-grained per-atom perspective. In this study, we propose an atom-level ProteinQA model, named Atom-ProteinQA, in which two innovative modules are designed to extract geometric and topological atom-level relationships respectively. Specifically, on the one hand, a geometric perception module exploits 3D sparse convolution to capture the geometric features of the input protein, generating fine-grained atom-level predictions. On the other hand, natural chemical bonds are utilized to construct an atom-level graph, then message passing from a topological perception module is applied to output residue-level predictions in parallel. Eventually, through a cross-model aggregation module, features from different modules mutually interact, enhancing performance on both the atom and residue levels.

RESULTS

Extensive experiments show that our proposed Atom-ProteinQA outperforms previous methods by a large margin, regardless of residue-level or atom-level assessment. Concretely, we achieved state-of-the-art performance on CATH-2084, Decoy-8000, public benchmarks CASP13 & CASP14, and the CAMEO.

AVAILABILITY

The repository of this project is released on: https://github.com/luyfcandy/Atom_ProteinQA.

Design of High Affinity Binders to Convex Protein Target Sites

While there has been progress in the de novo design of small globular miniproteins (50-65 residues) to bind to primarily concave regions of a target protein surface, computational design of minibinders to convex binding sites remains an outstanding challenge due to low level of overall shape complementarity. Here, we describe a general approach to generate computationally designed proteins which bind to convex target sites that employ geometrically matching concave scaffolds. We used this approach to design proteins binding to TGFβRII, CTLA-4 and PD-L1 which following experimental optimization have low nanomolar to picomolar affinities and potent biological activity. Co-crystal structures of the TGFβRII and CTLA-4 binders in complex with the receptors are in close agreement with the design models. Our approach provides a general route to generating very high affinity binders to convex protein target sites.

Application of Machine Learning in the Quantitative Analysis of the Surface Characteristics of Highly Abundant Cytoplasmic Proteins: Toward AI-Based Biomimetics

Proteins in the crowded environment of human cells have often been studied regarding nonspecific interactions, misfolding, and aggregation, which may cause cellular malfunction and disease. Specifically, proteins with high abundance are more susceptible to these issues due to the law of mass action. Therefore, the surfaces of highly abundant cytoplasmic (HAC) proteins directly exposed to the environment can exhibit specific physicochemical, structural, and geometrical characteristics that reduce nonspecific interactions and adapt to the environment. However, the quantitative relationships between the overall surface descriptors still need clarification. Here, we used machine learning to identify HAC proteins using hydrophobicity, charge, roughness, secondary structures, and B-factor from the protein surfaces and quantified the contribution of each descriptor. First, several supervised learning algorithms were compared to solve binary classification problems for the surfaces of HAC and extracellular proteins. Then, logistic regression was used for the feature importance analysis of descriptors considering model performance (80.2% accuracy and 87.6% AUC) and interpretability. The HAC proteins showed positive correlations with negatively and positively charged areas but negative correlations with hydrophobicity, the B-factor, the proportion of beta structures, roughness, and the proportion of disordered regions. Finally, the details of each descriptor could be explained concerning adaptative surface strategies of HAC proteins to regulate nonspecific interactions, protein folding, flexibility, stability, and adsorption. This study presented a novel approach using various surface descriptors to identify HAC proteins and provided quantitative design rules for the surfaces well-suited to human cellular crowded environments.

Protein structure accuracy estimation using geometry-complete perceptron networks

Estimating the accuracy of protein structural models is a critical task in protein bioinformatics. The need for robust methods in the estimation of protein model accuracy (EMA) is prevalent in the field of protein structure prediction, where computationally-predicted structures need to be screened rapidly for the reliability of the positions predicted for each of their amino acid residues and their overall quality. Current methods proposed for EMA are either coupled tightly to existing protein structure prediction methods or evaluate protein structures without sufficiently leveraging the rich, geometric information available in such structures to guide accuracy estimation. In this work, we propose a geometric message passing neural network referred to as the geometry-complete perceptron network for protein structure EMA (GCPNet-EMA), where we demonstrate through rigorous computational benchmarks that GCPNet-EMA's accuracy estimations are 47% faster and more than 10% (6%) more correlated with ground-truth measures of per-residue (per-target) structural accuracy compared to baseline state-of-the-art methods for tertiary (multimer) structure EMA including AlphaFold 2. The source code and data for GCPNet-EMA are available on GitHub, and a public web server implementation is freely available.

ProSTAGE: Predicting Effects of Mutations on Protein Stability by Using Protein Embeddings and Graph Convolutional Networks

Protein thermodynamic stability is essential to clarify the relationships among structure, function, and interaction. Therefore, developing a faster and more accurate method to predict the impact of the mutations on protein stability is helpful for protein design and understanding the phenotypic variation. Recent studies have shown that protein embedding will be particularly powerful at modeling sequence information with context dependence, such as subcellular localization, variant effect, and secondary structure prediction. Herein, we introduce a novel method, ProSTAGE, which is a deep learning method that fuses structure and sequence embedding to predict protein stability changes upon single point mutations. Our model combines graph-based techniques and language models to predict stability changes. Moreover, ProSTAGE is trained on a larger data set, which is almost twice as large as the most used S2648 data set. It consistently outperforms all existing state-of-the-art methods on mutation-affected problems as benchmarked on several independent data sets. The protein embedding as the prediction input achieves better results than the previous results, which shows the potential of protein language models in predicting the effect of mutations on proteins. ProSTAGE is implemented as a user-friendly web server.

MoDAFold: a strategy for predicting the structure of missense mutant protein based on AlphaFold2 and molecular dynamics

Protein structure prediction is a longstanding issue crucial for identifying new drug targets and providing a mechanistic understanding of protein functions. To enhance the progress in this field, a spectrum of computational methodologies has been cultivated. AlphaFold2 has exhibited exceptional precision in predicting wild-type protein structures, with performance exceeding that of other methods. However, predicting the structures of missense mutant proteins using AlphaFold2 remains challenging due to the intricate and substantial structural alterations caused by minor sequence variations in the mutant proteins. Molecular dynamics (MD) has been validated for precisely capturing changes in amino acid interactions attributed to protein mutations. Therefore, for the first time, a strategy entitled 'MoDAFold' was proposed to improve the accuracy and reliability of missense mutant protein structure prediction by combining AlphaFold2 with MD. Multiple case studies have confirmed the superior performance of MoDAFold compared to other methods, particularly AlphaFold2.

Assessment of the assessment-All about complexes

Predicting model quality is a fundamental component of any modeling procedure, and blind assessment of these methods constitutes a crucial aspect of the Critical Assessment of Protein Structure Prediction (CASP) experiment. Historically, the main focus was on assessing methods that predict global and per-residue accuracies in tertiary structure models. This focus shifted with the community's increased efforts in modeling complexes and assemblies. We asked the community to process the models from the CASP15 assembly category and provide estimates of the accuracy of the predicted quaternary structure, both globally and at the local interface level. Besides identifying remarkable accuracy of modeling groups in assessing their own predictions, we set up a benchmarking pipeline to highlight different aspects of quaternary structure models and introduced a simple consensus EMA method as baseline. While participating methods showed commendable performance, the baseline was difficult to surpass. It is important to point out that prediction performance varies for the individual CASP targets, highlighting potential areas of improvement and challenges ahead.

Phylogenetic and Protein Structure Analyses Provide Insight into the Evolution and Diversification of the CD36 Domain "Apex" among Scavenger Receptor Class B Proteins across Eukarya

The cluster of differentiation 36 (CD36) domain defines the characteristic ectodomain associated with class B scavenger receptor (SR-B) proteins. In bilaterians, SR-Bs play critical roles in diverse biological processes including innate immunity functions such as pathogen recognition and apoptotic cell clearance, as well as metabolic sensing associated with fatty acid uptake and cholesterol transport. Although previous studies suggest this protein family is ancient, SR-B diversity across Eukarya has not been robustly characterized. We analyzed SR-B homologs identified from the genomes and transcriptomes of 165 diverse eukaryotic species. The presence of highly conserved amino acid motifs across major eukaryotic supergroups supports the presence of a SR-B homolog in the last eukaryotic common ancestor. Our comparative analyses of SR-B protein structure identify the retention of a canonical asymmetric beta barrel tertiary structure within the CD36 ectodomain across Eukarya. We also identify multiple instances of independent lineage-specific sequence expansions in the apex region of the CD36 ectodomain-a region functionally associated with ligand-sensing. We hypothesize that a combination of both sequence expansion and structural variation in the CD36 apex region may reflect the evolution of SR-B ligand-sensing specificity between diverse eukaryotic clades.

Designing a novel and combinatorial multi-antigenic epitope-based vaccine "MarVax" against Marburg virus-a reverse vaccinology and immunoinformatics approach

CONTEXT

Marburg virus (MARV) is a member of the Filoviridae family and causes Marburg virus disease (MVD) among humans and primates. With fatality rates going up to 88%, there is currently no commercialized cure or vaccine to combat the infection. The National Institute of Allergy and Infectious Diseases (NIAID) classified MARV as priority pathogen A, which presages the need for a vaccine candidate which can provide stable, long-term adaptive immunity. The surface glycoprotein (GP) and fusion protein (FP) mediate the adherence, fusion, and entry of the virus into the host cell via the TIM-I receptor. Being important antigenic determinants, studies reveal that GP and FP are prone to evolutionary mutations, underscoring the requirement of a vaccine construct capable of eliciting a robust and sustained immune response. In this computational study, a reverse vaccinology approach was employed to design a combinatorial vaccine from conserved and antigenic epitopes of essential viral proteins of MARV, namely GP, VP24, VP30, VP35, and VP40 along with an endogenous protein large polymerase (L).

METHODS

Epitopes for T-cell and B-cell were predicted using TepiTool and ElliPro, respectively. The surface-exposed TLRs like TLR2, TLR4, and TLR5 were used to screen high-binding affinity epitopes using the protein-peptide docking platform MdockPeP. The best binding epitopes were selected and assembled with linkers to design a recombinant multi-epitope vaccine construct which was then modeled in Robetta. The in silico biophysical and biochemical analyses of the recombinant vaccine were performed. The docking and MD simulation of the vaccine using WebGro and CABS-Flex against TLRs support the stable binding of vaccine candidates. A virtual immune simulation to check the immediate and long-term immunogenicity was carried out using the C-ImmSim server.

RESULTS

The biochemical characteristics and docking studies with MD simulation establish the recombinant protein vaccine construct MarVax as a stable, antigenic, and potent vaccine molecule. Immune simulation studies reveal 1-year passive immunity which needs to be validated by in vivo studies.

Protein quality assessment with a loss function designed for high-quality decoys

Motivation: The prediction of a protein 3D structure is essential for understanding protein function, drug discovery, and disease mechanisms; with the advent of methods like AlphaFold that are capable of producing very high-quality decoys, ensuring the quality of those decoys can provide further confidence in the accuracy of their predictions. Results: In this work, we describe Qϵ, a graph convolutional network (GCN) that utilizes a minimal set of atom and residue features as inputs to predict the global distance test total score (GDTTS) and local distance difference test (lDDT) score of a decoy. To improve the model's performance, we introduce a novel loss function based on the ϵ-insensitive loss function used for SVM regression. This loss function is specifically designed for evaluating the characteristics of the quality assessment problem and provides predictions with improved accuracy over standard loss functions used for this task. Despite using only a minimal set of features, it matches the performance of recent state-of-the-art methods like DeepUMQA. Availability: The code for Qϵ is available at https://github.com/soumyadip1997/qepsilon.

DeepUMQA3: a web server for accurate assessment of interface residue accuracy in protein complexes

MOTIVATION

Model quality assessment is a crucial part of protein structure prediction and a gateway to proper usage of models in biomedical applications. Many methods have been proposed for assessing the quality of structural models of protein monomers, but few methods for evaluating protein complex models. As protein complex structure prediction becomes a new challenge, there is an urgent need for model quality assessment methods that can accurately assess the accuracy of interface residues of complex structures.

RESULTS

Here, we present DeepUMQA3, a web server for evaluating the accuracy of interface residues of protein complex structures using deep neural networks. For an input complex structure, features are extracted from three levels of overall complex, intra-monomer, and inter-monomer, and an improved deep residual neural network is used to predict per-residue lDDT and interface residue accuracy. DeepUMQA3 ranks first in the blind test of interface residue accuracy estimation in CASP15, with Pearson, Spearman, and AUC of 0.564, 0.535, and 0.755 under the lDDT measurement, which are 17.6%, 23.6%, and 10.9% higher than the second best method, respectively. DeepUMQA3 can also assess the accuracy of all residues in the entire complex and distinguish high- and low-precision residues.

AVAILABILITY AND IMPLEMENTATION

The web sever of DeepUMQA3 are freely available at http://zhanglab-bioinf.com/DeepUMQA_server/.

AI-driven pan-proteome analyses reveal insights into the biohydrometallurgical properties of Acidithiobacillia

Microorganism-mediated biohydrometallurgy, a sustainable approach for metal recovery from ores, relies on the metabolic activity of acidophilic bacteria. Acidithiobacillia with sulfur/iron-oxidizing capacities are extensively studied and applied in biohydrometallurgy-related processes. However, only 14 distinct proteins from Acidithiobacillia have experimentally determined structures currently available. This significantly hampers in-depth investigations of Acidithiobacillia's structure-based biological mechanisms pertaining to its relevant biohydrometallurgical processes. To address this issue, we employed a state-of-the-art artificial intelligence (AI)-driven approach, with a median model confidence of 0.80, to perform high-quality full-chain structure predictions on the pan-proteome (10,458 proteins) of the type strain Acidithiobacillia. Additionally, we conducted various case studies on de novo protein structural prediction, including sulfate transporter and iron oxidase, to demonstrate how accurate structure predictions and gene co-occurrence networks can contribute to the development of mechanistic insights and hypotheses regarding sulfur and iron utilization proteins. Furthermore, for the unannotated proteins that constitute 35.8% of the Acidithiobacillia proteome, we employed the deep-learning algorithm DeepFRI to make structure-based functional predictions. As a result, we successfully obtained gene ontology (GO) terms for 93.6% of these previously unknown proteins. This study has a significant impact on improving protein structure and function predictions, as well as developing state-of-the-art techniques for high-throughput analysis of large proteomic data.

Biochemical evidence that the whole compartment activity behavior of GAPDH differs between the cytoplasm and nucleus

Some metabolic enzymes normally occur in the nucleus and cytoplasm. These compartments differ in molecular composition. Since post-translational modification and interaction with allosteric effectors can tune enzyme activity, it follows that the behavior of an enzyme as a catalyst may differ between the cytoplasm and nucleus. We explored this possibility for the glycolytic enzyme glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Homogenates of pristine nuclei and cytoplasms isolated from Xenopus laevis oocytes were used for whole compartment activity profiling in a near-physiological buffer. Titrations of NAD+ revealed similar whole compartment activity profiles for GAPDH in nuclear and cytoplasmic homogenates. Surprisingly however GAPDH in these compartments did not have the same behavior in assays of the dependence of initial velocity (v0) on G3P concentration. First, the peak v0 for nuclear GAPDH was up to 2.5-fold higher than the peak for cytoplasmic GAPDH. Second, while Michaelis Menten-like behavior was observed in all assays of cytoplasm, the v0 versus [G3P] plots for nuclear GAPDH typically exhibited a non-Michaelis Menten (sigmoidal) profile. Apparent Km and Vmax (G3P) values for nuclear GAPDH activity were highly variable, even between replicates of the same sample. Possible sources of this variability include in vitro processing of a metabolite that allosterically regulates GAPDH, turnover of a post-translational modification of the enzyme, and fluctuation of the state of interaction of GAPDH with other proteins. Collectively these findings are consistent with the hypothesis that the environment of the nucleus is distinct from the environment of the cytoplasm with regard to GAPDH activity and its modulation. This finding warrants further comparison of the regulation of nuclear and cytoplasmic GAPDH, as well as whole compartment activity profiling of other enzymes of metabolism with cytosolic and nuclear pools.

DUF2285 is a novel helix-turn-helix domain variant that orchestrates both activation and antiactivation of conjugative element transfer in proteobacteria

Horizontal gene transfer is tightly regulated in bacteria. Often only a fraction of cells become donors even when regulation of horizontal transfer is coordinated at the cell population level by quorum sensing. Here, we reveal the widespread 'domain of unknown function' DUF2285 represents an 'extended-turn' variant of the helix-turn-helix domain that participates in both transcriptional activation and antiactivation to initiate or inhibit horizontal gene transfer. Transfer of the integrative and conjugative element ICEMlSymR7A is controlled by the DUF2285-containing transcriptional activator FseA. One side of the DUF2285 domain of FseA has a positively charged surface which is required for DNA binding, while the opposite side makes critical interdomain contacts with the N-terminal FseA DUF6499 domain. The QseM protein is an antiactivator of FseA and is composed of a DUF2285 domain with a negative surface charge. While QseM lacks the DUF6499 domain, it can bind the FseA DUF6499 domain and prevent transcriptional activation by FseA. DUF2285-domain proteins are encoded on mobile elements throughout the proteobacteria, suggesting regulation of gene transfer by DUF2285 domains is a widespread phenomenon. These findings provide a striking example of how antagonistic domain paralogues have evolved to provide robust molecular control over the initiation of horizontal gene transfer.

Residue-level error detection in cryoelectron microscopy models

Building accurate protein models into moderate resolution (3-5 Å) cryoelectron microscopy (cryo-EM) maps is challenging and error prone. We have developed MEDIC (Model Error Detection in Cryo-EM), a robust statistical model that identifies local backbone errors in protein structures built into cryo-EM maps by combining local fit-to-density with deep-learning-derived structural information. MEDIC is validated on a set of 28 structures that were subsequently solved to higher resolutions, where we identify the differences between low- and high-resolution structures with 68% precision and 60% recall. We additionally use this model to fix over 100 errors in 12 deposited structures and to identify errors in 4 refined AlphaFold predictions with 80% precision and 60% recall. As modelers more frequently use deep learning predictions as a starting point for refinement and rebuilding, MEDIC's ability to handle errors in structures derived from hand-building and machine learning methods makes it a powerful tool for structural biologists.

DeepAlgPro: an interpretable deep neural network model for predicting allergenic proteins

Allergies have become an emerging public health problem worldwide. The most effective way to prevent allergies is to find the causative allergen at the source and avoid re-exposure. However, most of the current computational methods used to identify allergens were based on homology or conventional machine learning methods, which were inefficient and still had room to be improved for the detection of allergens with low homology. In addition, few methods based on deep learning were reported, although deep learning has been successfully applied to several tasks in protein sequence analysis. In the present work, a deep neural network-based model, called DeepAlgPro, was proposed to identify allergens. We showed its great accuracy and applicability to large-scale forecasts by comparing it to other available tools. Additionally, we used ablation experiments to demonstrate the critical importance of the convolutional module in our model. Moreover, further analyses showed that epitope features contributed to model decision-making, thus improving the model's interpretability. Finally, we found that DeepAlgPro was capable of detecting potential new allergens. Overall, DeepAlgPro can serve as powerful software for identifying allergens.

Prediction of CD44 Structure by Deep Learning-Based Protein Modeling

CD44 is a cell surface glycoprotein transmembrane receptor that is involved in cell-cell and cell-matrix interactions. It crucially associates with several molecules composing the extracellular matrix, the main one of which is hyaluronic acid. It is ubiquitously expressed in various types of cells and is involved in the regulation of important signaling pathways, thus playing a key role in several physiological and pathological processes. Structural information about CD44 is, therefore, fundamental for understanding the mechanism of action of this receptor and developing effective treatments against its aberrant expression and dysregulation frequently associated with pathological conditions. To date, only the structure of the hyaluronan-binding domain (HABD) of CD44 has been experimentally determined. To elucidate the nature of CD44s, the most frequently expressed isoform, we employed the recently developed deep-learning-based tools D-I-TASSER, AlphaFold2, and RoseTTAFold for an initial structural prediction of the full-length receptor, accompanied by molecular dynamics simulations on the most promising model. All three approaches correctly predicted the HABD, with AlphaFold2 outperforming D-I-TASSER and RoseTTAFold in the structural comparison with the crystallographic HABD structure and confidence in predicting the transmembrane helix. Low confidence regions were also predicted, which largely corresponded to the disordered regions of CD44s. These regions allow the receptor to perform its unconventional activity.

GraphGPSM: a global scoring model for protein structure using graph neural networks

The scoring models used for protein structure modeling and ranking are mainly divided into unified field and protein-specific scoring functions. Although protein structure prediction has made tremendous progress since CASP14, the modeling accuracy still cannot meet the requirements to a certain extent. Especially, accurate modeling of multi-domain and orphan proteins remains a challenge. Therefore, an accurate and efficient protein scoring model should be developed urgently to guide the protein structure folding or ranking through deep learning. In this work, we propose a protein structure global scoring model based on equivariant graph neural network (EGNN), named GraphGPSM, to guide protein structure modeling and ranking. We construct an EGNN architecture, and a message passing mechanism is designed to update and transmit information between nodes and edges of the graph. Finally, the global score of the protein model is output through a multilayer perceptron. Residue-level ultrafast shape recognition is used to describe the relationship between residues and the overall structure topology, and distance and direction encoded by Gaussian radial basis functions are designed to represent the overall topology of the protein backbone. These two features are combined with Rosetta energy terms, backbone dihedral angles and inter-residue distance and orientations to represent the protein model and embedded into the nodes and edges of the graph neural network. The experimental results on the CASP13, CASP14 and CAMEO test sets show that the scores of our developed GraphGPSM have a strong correlation with the TM-score of the models, which are significantly better than those of the unified field score function REF2015 and the state-of-the-art local lDDT-based scoring models ModFOLD8, ProQ3D and DeepAccNet, etc. The modeling experimental results on 484 test proteins demonstrate that GraphGPSM can greatly improve the modeling accuracy. GraphGPSM is further used to model 35 orphan proteins and 57 multi-domain proteins. The results show that the average TM-score of the models predicted by GraphGPSM is 13.2 and 7.1% higher than that of the models predicted by AlphaFold2. GraphGPSM also participates in CASP15 and achieves competitive performance in global accuracy estimation.

CD28 and 41BB Costimulatory Domains Alone or in Combination Differentially Influence Cell Surface Dynamics and Organization of Chimeric Antigen Receptors and Early Activation of CAR T Cells

Chimeric antigen receptor (CAR)-modified T cells brought a paradigm shift in the treatment of chemotherapy-resistant lymphomas. Conversely, clinical experience with CAR T cells targeting solid tumors has been disheartening, indicating the necessity of their molecular-level optimization. While incorporating CD28 or 41BB costimulatory domains into CARs in addition to the CD3z signaling domain improved the long-term efficacy of T cell products, their influence on early tumor engagement has yet to be elucidated. We studied the antigen-independent self-association and membrane diffusion kinetics of first- (.z), second- (CD28.z, 41BB.z), and third- (CD28.41BB.z) generation HER2-specific CARs in the resting T cell membrane using super-resolution AiryScan microscopy and fluorescence correlation spectroscopy, in correlation with RoseTTAFold-based structure prediction and assessment of oligomerization in native Western blot. While .z and CD28.z CARs formed large, high-density submicron clusters of dimers, 41BB-containing CARs formed higher oligomers that assembled into smaller but more numerous membrane clusters. The first-, second-, and third-generation CARs showed progressively increasing lateral diffusion as the distance of their CD3z domain from the membrane plane increased. Confocal microscopy analysis of immunological synapses showed that both small clusters of highly mobile CD28.41BB.z and large clusters of less mobile .z CAR induced more efficient CD3ζ and pLck phosphorylation than CD28.z or 41BB.z CARs of intermediate mobility. However, electric cell-substrate impedance sensing revealed that the CD28.41BB.z CAR performs worst in sequential short-term elimination of adherent tumor cells, while the .z CAR is superior to all others. We conclude that the molecular structure, membrane organization, and mobility of CARs are critical design parameters that can predict the development of an effective immune synapse. Therefore, they need to be taken into account alongside the long-term biological effects of costimulatory domains to achieve an optimal therapeutic effect.

Improving de novo protein binder design with deep learning

Recently it has become possible to de novo design high affinity protein binding proteins from target structural information alone. There is, however, considerable room for improvement as the overall design success rate is low. Here, we explore the augmentation of energy-based protein binder design using deep learning. We find that using AlphaFold2 or RoseTTAFold to assess the probability that a designed sequence adopts the designed monomer structure, and the probability that this structure binds the target as designed, increases design success rates nearly 10-fold. We find further that sequence design using ProteinMPNN rather than Rosetta considerably increases computational efficiency.

Atomic protein structure refinement using all-atom graph representations and SE(3)-equivariant graph transformer

MOTIVATION

The state-of-art protein structure prediction methods such as AlphaFold are being widely used to predict structures of uncharacterized proteins in biomedical research. There is a significant need to further improve the quality and nativeness of the predicted structures to enhance their usability. In this work, we develop ATOMRefine, a deep learning-based, end-to-end, all-atom protein structural model refinement method. It uses a SE(3)-equivariant graph transformer network to directly refine protein atomic coordinates in a predicted tertiary structure represented as a molecular graph.

RESULTS

The method is first trained and tested on the structural models in AlphaFoldDB whose experimental structures are known, and then blindly tested on 69 CASP14 regular targets and 7 CASP14 refinement targets. ATOMRefine improves the quality of both backbone atoms and all-atom conformation of the initial structural models generated by AlphaFold. It also performs better than two state-of-the-art refinement methods in multiple evaluation metrics including an all-atom model quality score-the MolProbity score based on the analysis of all-atom contacts, bond length, atom clashes, torsion angles, and side-chain rotamers. As ATOMRefine can refine a protein structure quickly, it provides a viable, fast solution for improving protein geometry and fixing structural errors of predicted structures through direct coordinate refinement.

AVAILABILITY AND IMPLEMENTATION

The source code of ATOMRefine is available in the GitHub repository (https://github.com/BioinfoMachineLearning/ATOMRefine). All the required data for training and testing are available at https://doi.org/10.5281/zenodo.6944368.

Designing artificial pathways for improving chemical production

Metabolic engineering exploits manipulation of catalytic and regulatory elements to improve a specific function of the host cell, often the synthesis of interesting chemicals. Although naturally occurring pathways are significant resources for metabolic engineering, these pathways are frequently inefficient and suffer from a series of inherent drawbacks. Designing artificial pathways in a rational manner provides a promising alternative for chemicals production. However, the entry barrier of designing artificial pathway is relatively high, which requires researchers a comprehensive and deep understanding of physical, chemical and biological principles. On the other hand, the designed artificial pathways frequently suffer from low efficiencies, which impair their further applications in host cells. Here, we illustrate the concept and basic workflow of retrobiosynthesis in designing artificial pathways, as well as the most currently used methods including the knowledge- and computer-based approaches. Then, we discuss how to obtain desired enzymes for novel biochemistries, and how to trim the initially designed artificial pathways for further improving their functionalities. Finally, we summarize the current applications of artificial pathways from feedstocks utilization to various products synthesis, as well as our future perspectives on designing artificial pathways.

Do "Newly Born" orphan proteins resemble "Never Born" proteins? A study using three deep learning algorithms

"Newly Born" proteins, devoid of detectable homology to any other proteins, known as orphan proteins, occur in a single species or within a taxonomically restricted gene family. They are generated by the expression of novel open reading frames, and appear throughout evolution. We were curious if three recently developed programs for predicting protein structures, namely, AlphaFold2, RoseTTAFold, and ESMFold, might be of value for comparison of such "Newly Born" proteins to random polypeptides with amino acid content similar to that of native proteins, which have been called "Never Born" proteins. The programs were used to compare the structures of two sets of "Never Born" proteins that had been expressed-Group 1, which had been shown experimentally to possess substantial secondary structure, and Group 3, which had been shown to be intrinsically disordered. Overall, although the models generated were scored as being of low quality, they nevertheless revealed some general principles. Specifically, all four members of Group 1 were predicted to be compact by all three algorithms, in agreement with the experimental data, whereas the members of Group 3 were predicted to be very extended, as would be expected for intrinsically disordered proteins, again consistent with the experimental data. These predicted differences were shown to be statistically significant by comparing their accessible surface areas. The three programs were then used to predict the structures of three orphan proteins whose crystal structures had been solved, two of which display novel folds. Surprisingly, only for the protein which did not have a novel fold, and was taxonomically restricted, rather than being a true orphan, did all three algorithms predict very similar, high-quality structures, closely resembling the crystal structure. Finally, they were used to predict the structures of seven orphan proteins with well-identified biological functions, whose 3D structures are not known. Two proteins, which were predicted to be disordered based on their sequences, are predicted by all three structure algorithms to be extended structures. The other five were predicted to be compact structures with only two exceptions in the case of AlphaFold2. All three prediction algorithms make remarkably similar and high-quality predictions for one large protein, HCO_11565, from a nematode. It is conjectured that this is due to many homologs in the taxonomically restricted family of which it is a member, and to the fact that the Dali server revealed several nonrelated proteins with similar folds. An animated Interactive 3D Complement (I3DC) is available in Proteopedia at http://proteopedia.org/w/Journal:Proteins:3.

Using AlphaFold Predictions in Viral Research

Elucidation of the tertiary structure of proteins is an important task for biological and medical studies. AlphaFold, a modern deep-learning algorithm, enables the prediction of protein structure to a high level of accuracy. It has been applied in numerous studies in various areas of biology and medicine. Viruses are biological entities infecting eukaryotic and procaryotic organisms. They can pose a danger for humans and economically significant animals and plants, but they can also be useful for biological control, suppressing populations of pests and pathogens. AlphaFold can be used for studies of molecular mechanisms of viral infection to facilitate several activities, including drug design. Computational prediction and analysis of the structure of bacteriophage receptor-binding proteins can contribute to more efficient phage therapy. In addition, AlphaFold predictions can be used for the discovery of enzymes of bacteriophage origin that are able to degrade the cell wall of bacterial pathogens. The use of AlphaFold can assist fundamental viral research, including evolutionary studies. The ongoing development and improvement of AlphaFold can ensure that its contribution to the study of viral proteins will be significant in the future.

Statistical genetics in and out of quasi-linkage equilibrium

This review is about statistical genetics, an interdisciplinary topic between statistical physics and population biology. The focus is on the phase ofquasi-linkage equilibrium(QLE). Our goals here are to clarify under which conditions the QLE phase can be expected to hold in population biology and how the stability of the QLE phase is lost. The QLE state, which has many similarities to a thermal equilibrium state in statistical mechanics, was discovered by M Kimura for a two-locus two-allele model, and was extended and generalized to the global genome scale byNeher&Shraiman (2011). What we will refer to as the Kimura-Neher-Shraiman theory describes a population evolving due to the mutations, recombination, natural selection and possibly genetic drift. A QLE phase exists at sufficiently high recombination rate (r) and/or mutation ratesµwith respect to selection strength. We show how in QLE it is possible to infer the epistatic parameters of the fitness function from the knowledge of the (dynamical) distribution of genotypes in a population. We further consider the breakdown of the QLE regime for high enough selection strength. We review recent results for the selection-mutation and selection-recombination dynamics. Finally, we identify and characterize a new phase which we call the non-random coexistence where variability persists in the population without either fixating or disappearing.

Lignin deconstruction by anaerobic fungi

Lignocellulose forms plant cell walls, and its three constituent polymers, cellulose, hemicellulose and lignin, represent the largest renewable organic carbon pool in the terrestrial biosphere. Insights into biological lignocellulose deconstruction inform understandings of global carbon sequestration dynamics and provide inspiration for biotechnologies seeking to address the current climate crisis by producing renewable chemicals from plant biomass. Organisms in diverse environments disassemble lignocellulose, and carbohydrate degradation processes are well defined, but biological lignin deconstruction is described only in aerobic systems. It is currently unclear whether anaerobic lignin deconstruction is impossible because of biochemical constraints or, alternatively, has not yet been measured. We applied whole cell-wall nuclear magnetic resonance, gel-permeation chromatography and transcriptome sequencing to interrogate the apparent paradox that anaerobic fungi (Neocallimastigomycetes), well-documented lignocellulose degradation specialists, are unable to modify lignin. We find that Neocallimastigomycetes anaerobically break chemical bonds in grass and hardwood lignins, and we further associate upregulated gene products with the observed lignocellulose deconstruction. These findings alter perceptions of lignin deconstruction by anaerobes and provide opportunities to advance decarbonization biotechnologies that depend on depolymerizing lignocellulose.