Evaluation of AlphaFold Antibody-Antigen Modeling with Implications for Improving Predictive Accuracy

Prediction of protein structure and AI

AlphaFold, an artificial intelligence (AI)-based tool for predicting the 3D structure of proteins, is now widely recognized for its high accuracy and versatility in the folding of human proteins. AlphaFold is useful for understanding structure-function relationships from protein 3D structure models and can serve as a template or a reference for experimental structural analysis including X-ray crystallography, NMR and cryo-EM analysis. Its use is expanding among researchers, not only in structural biology but also in other research fields. Researchers are currently exploring the full potential of AlphaFold-generated protein models. Predicting disease severity caused by missense mutations is one such application. This article provides an overview of the 3D structural modeling of AlphaFold based on deep learning techniques and highlights the challenges in predicting the pathogenicity of missense mutations.

Integrating Dynamic Network Analysis with AI for Enhanced Epitope Prediction in PD-L1:Affibody Interactions

Understanding binding epitopes involved in protein-protein interactions and accurately determining their structure are long-standing goals with broad applicability in industry and biomedicine. Although various experimental methods for binding epitope determination exist, these approaches are typically low throughput and cost-intensive. Computational methods have potential to accelerate epitope predictions; however, recently developed artificial intelligence (AI)-based methods frequently fail to predict epitopes of synthetic binding domains with few natural homologues. Here we have developed an integrated method employing generalized-correlation-based dynamic network analysis on multiple molecular dynamics (MD) trajectories, initiated from AlphaFold2Multimer structures, to unravel the structure and binding epitope of the therapeutic PD-L1:Affibody complex. Both AlphaFold2 and conventional molecular dynamics trajectory analysis were ineffective in distinguishing between two proposed binding models, parallel and perpendicular. However, our integrated approach, utilizing dynamic network analysis, demonstrated that the perpendicular mode was significantly more stable. These predictions were validated using a suite of experimental epitope mapping protocols, including cross-linking mass spectrometry and next-generation sequencing-based deep mutational scanning. Conversely, AlphaFold3 failed to predict a structure bound in the perpendicular pose, highlighting the necessity for exploratory research in the search for binding epitopes and challenging the notion that AI-generated protein structures can be accepted without scrutiny. Our research underscores the potential of employing dynamic network analysis to enhance AI-based structure predictions for more accurate identification of protein-protein interaction interfaces.

Molecular Modeling Methods in the Development of Affine and Specific Protein-Binding Agents

High-affinity and specific agents are widely applied in various areas, including diagnostics, scientific research, and disease therapy (as drugs and drug delivery systems). It takes significant time to develop them. For this reason, development of high-affinity agents extensively utilizes computer methods at various stages for the analysis and modeling of these molecules. The review describes the main affinity and specific agents, such as monoclonal antibodies and their fragments, antibody mimetics, aptamers, and molecularly imprinted polymers. The methods of their obtaining as well as their main advantages and disadvantages are briefly described, with special attention focused on the molecular modeling methods used for their analysis and development.

Septins function in exocytosis via physical interactions with the exocyst complex in fission yeast cytokinesis

Septins can function as scaffolds for protein recruitment, membrane-bound diffusion barriers, or membrane curvature sensors. Septins are important for cytokinesis, but their exact roles are still obscure. In fission yeast, four septins (Spn1 to Spn4) accumulate at the rim of the division plane as rings. The octameric exocyst complex, which tethers exocytic vesicles to the plasma membrane, exhibits a similar localization and is essential for plasma membrane deposition during cytokinesis. Without septins, the exocyst spreads across the division plane but absent from the rim during septum formation. These results suggest that septins and the exocyst physically interact for proper localization. Indeed, we predicted six pairs of direct interactions between septin and exocyst subunits by AlphaFold2 ColabFold, most of them are confirmed by co-immunoprecipitation and yeast two-hybrid assays. Exocyst mislocalization results in mistargeting of secretory vesicles and their cargos, which leads to cell-separation delay in septin mutants. Our results indicate that septins guide the targeting of exocyst complex on the plasma membrane for vesicle tethering during cytokinesis through direct physical interactions.

Atomically accurate de novo design of single-domain antibodies

Despite the central role that antibodies play in modern medicine, there is currently no way to rationally design novel antibodies to bind a specific epitope on a target. Instead, antibody discovery currently involves time-consuming immunization of an animal or library screening approaches. Here we demonstrate that a fine-tuned RFdiffusion network is capable of designing de novo antibody variable heavy chains (VHH's) that bind user-specified epitopes. We experimentally confirm binders to four disease-relevant epitopes, and the cryo-EM structure of a designed VHH bound to influenza hemagglutinin is nearly identical to the design model both in the configuration of the CDR loops and the overall binding pose.

Integrating Dynamic Network Analysis with AI for Enhanced Epitope Prediction in PD-L1:Affibody Interactions

Understanding binding epitopes involved in protein-protein interactions and accurately determining their structure is a long standing goal with broad applicability in industry and biomedicine. Although various experimental methods for binding epitope determination exist, these approaches are typically low throughput and cost intensive. Computational methods have potential to accelerate epitope predictions, however, recently developed artificial intelligence (AI)-based methods frequently fail to predict epitopes of synthetic binding domains with few natural homologs. Here we have developed an integrated method employing generalized-correlation-based dynamic network analysis on multiple molecular dynamics (MD) trajectories, initiated from AlphaFold2 Multimer structures, to unravel the structure and binding epitope of the therapeutic PD-L1:Affibody complex. Both AlphaFold2 and conventional molecular dynamics trajectory analysis alone each proved ineffectual in differentiating between two putative binding models referred to as parallel and perpendicular. However, our integrated approach based on dynamic network analysis showed that the perpendicular mode was significantly more stable. These predictions were validated using a suite of experimental epitope mapping protocols including cross linking mass spectrometry and next-generation sequencing-based deep mutational scanning. Our research highlights the potential of deploying dynamic network analysis to refine AI-based structure predictions for precise predictions of protein-protein interaction interfaces.

Applications and challenges in designing VHH-based bispecific antibodies: leveraging machine learning solutions

The development of bispecific antibodies that bind at least two different targets relies on bringing together multiple binding domains with different binding properties and biophysical characteristics to produce a drug-like therapeutic. These building blocks play an important role in the overall quality of the molecule and can influence many important aspects from potency and specificity to stability and half-life. Single-domain antibodies, particularly camelid-derived variable heavy domain of heavy chain (VHH) antibodies, are becoming an increasingly popular choice for bispecific construction due to their single-domain modularity, favorable biophysical properties, and potential to work in multiple antibody formats. Here, we review the use of VHH domains as building blocks in the construction of multispecific antibodies and the challenges in creating optimized molecules. In addition to exploring traditional approaches to VHH development, we review the integration of machine learning techniques at various stages of the process. Specifically, the utilization of machine learning for structural prediction, lead identification, lead optimization, and humanization of VHH antibodies.

Experimental structures of antibody/MHC-I complexes reveal details of epitopes overlooked by computational prediction

Monoclonal antibodies (mAb) to major histocompatibility complex class I (MHC-I) molecules have proved to be crucial reagents for tissue typing and fundamental studies of immune recognition. To augment our understanding of epitopic sites seen by a set of anti-MHC-I mAb, we determined X-ray crystal structures of four complexes of anti-MHC-I antigen-binding fragments (Fab) bound to peptide/MHC-I/β2m (pMHC-I). An anti-H2-Dd mAb, two anti-MHC-I α3 domain mAb, and an anti-β2-microglobulin (β2m) mAb bind pMHC-I at sites consistent with earlier mutational and functional experiments, and the structures explain allelomorph specificity. Comparison of the experimentally determined structures with computationally derived models using AlphaFold Multimer (AF-M) showed that although predictions of the individual pMHC-I heterodimers were quite acceptable, the computational models failed to properly identify the docking sites of the mAb on pMHC-I. The experimental and predicted structures provide insight into strengths and weaknesses of purely computational approaches and suggest areas that merit additional attention.

Reliable protein-protein docking with AlphaFold, Rosetta, and replica-exchange

Despite the recent breakthrough of AlphaFold (AF) in the field of protein sequence-to-structure prediction, modeling protein interfaces and predicting protein complex structures remains challenging, especially when there is a significant conformational change in one or both binding partners. Prior studies have demonstrated that AF-multimer (AFm) can predict accurate protein complexes in only up to 43% of cases.1 In this work, we combine AlphaFold as a structural template generator with a physics-based replica exchange docking algorithm. Using a curated collection of 254 available protein targets with both unbound and bound structures, we first demonstrate that AlphaFold confidence measures (pLDDT) can be repurposed for estimating protein flexibility and docking accuracy for multimers. We incorporate these metrics within our ReplicaDock 2.0 protocol2 to complete a robust in-silico pipeline for accurate protein complex structure prediction. AlphaRED (AlphaFold-initiated Replica Exchange Docking) successfully docks failed AF predictions including 97 failure cases in Docking Benchmark Set 5.5. AlphaRED generates CAPRI acceptable-quality or better predictions for 66% of benchmark targets. Further, on a subset of antigen-antibody targets, which is challenging for AFm (19% success rate), AlphaRED demonstrates a success rate of 51%. This new strategy demonstrates the success possible by integrating deep-learning based architectures trained on evolutionary information with physics-based enhanced sampling. The pipeline is available at github.com/Graylab/AlphaRED.