Deploying synthetic coevolution and machine learning to engineer protein-protein interactions

Synthetic biology and artificial intelligence in crop improvement

Synthetic biology plays a pivotal role in improving crop traits and increasing bioproduction through the use of engineering principles that purposefully modify plants through "design, build, test, and learn" cycles, ultimately resulting in improved bioproduction based on an input genetic circuit (DNA, RNA, and proteins). Crop synthetic biology is a new tool that uses circular principles to redesign and create innovative biological components, devices, and systems to enhance yields, nutrient absorption, resilience, and nutritional quality. In the digital age, artificial intelligence (AI) has demonstrated great strengths in design and learning. The application of AI has become an irreversible trend, with particularly remarkable potential for use in crop breeding. However, there has not yet been a systematic review of AI-driven synthetic biology pathways for plant engineering. In this review, we explore the fundamental engineering principles used in crop synthetic biology and their applications for crop improvement. We discuss approaches to genetic circuit design, including gene editing, synthetic nucleic acid and protein technologies, multi-omics analysis, genomic selection, directed protein engineering, and AI. We then outline strategies for the development of crops with higher photosynthetic efficiency, reshaped plant architecture, modified metabolic pathways, and improved environmental adaptability and nutrient absorption; the establishment of trait networks; and the construction of crop factories. We propose the development of SMART (self-monitoring, adapted, and responsive technology) crops through AI-empowered synthetic biotechnology. Finally, we address challenges associated with the development of synthetic biology and offer potential solutions for crop improvement.

Combining Directed Evolution with Machine Learning Enables Accurate Genotype-to-Phenotype Predictions

Linking sequence variation to phenotypic effects is critical for efficient exploitation of large genomic datasets. Here we present a novel approach combining directed evolution with protein language modeling to characterize naturally-evolved variants of a rice immune receptor. Using high-throughput directed evolution, we engineered the rice immune receptor Pik-1 to bind and recognize the fungal proteins Avr-PikC and Avr-PikF, which evade detection by currently characterized Pik-1 alleles. A protein language model was fine-tuned on this data to correlate sequence variation with ligand binding behavior. This modeling was then used to characterize Pik-1 variants found in the 3,000 Rice Genomes Project dataset. Two variants scored highly for binding against Avr-PikC, and in vitro analyses confirmed their improved ligand binding over the wild-type Pik-1 receptor. Overall, this machine learning approach identified promising sources of disease resistance in rice and shows potential utility for exploring the phenotypic variation of other proteins of interest.

Conformation Influences Biological Fates of Peptide-Based Nanofilaments by Modulating Protein Adsorption and Interfilament Entanglement

Filamentous structures exert biological functions mediated by multivalent interactions with their counterparts in sharp contrast with spherical ones. The physicochemical properties and unique behaviors of nanofilaments that are associated with multivalent interaction with protein are poorly understood. Here, peptide-based nanofilaments containing different homotetrapeptidic inserts are reported and their protein adsorption and biological fates are tested. By altering the homotetrapeptides, different peptidic conformations are imposed within the nanofilaments, which result in notable differences in the density of the intermolecular hydrogen bond, determining the amount of adsorbed proteins. The adsorbed proteins can further induce interfilament entanglement of different degrees and patterns, which influences biodistribution and phagocytosis. The nanofilaments with tetrahydroxyproline segment exhibit diminish interfilament entanglement, phagocytosis, and improve circulation, biodistribution, and antitumor efficacy. These findings can deepen the understanding of nanofilament-protein interactions and filament-filament interactions as in the case of amyloid-β plaque, and facilitate the rational design of nanofilaments through peptide conformation control for chemical engineering and anticancer drug delivery.

TPepPro: a deep learning model for predicting peptide-protein interactions

MOTIVATION

Peptides and their derivatives hold potential as therapeutic agents. The rising interest in developing peptide drugs is evidenced by increasing approval rates by the FDA of USA. To identify the most potential peptides, study on peptide-protein interactions (PepPIs) presents a very important approach but poses considerable technical challenges. In experimental aspects, the transient nature of PepPIs and the high flexibility of peptides contribute to elevated costs and inefficiency. Traditional docking and molecular dynamics simulation methods require substantial computational resources, and the predictive accuracy of their results remain unsatisfactory.

RESULTS

To address this gap, we proposed TPepPro, a Transformer-based model for PepPI prediction. We trained TPepPro on a dataset of 19,187 pairs of peptide-protein complexes with both sequential and structural features. TPepPro utilizes a strategy that combines local protein sequence feature extraction with global protein structure feature extraction. Moreover, TPepPro optimizes the architecture of structural featuring neural network in BN-ReLU arrangement, which notably reduced the amount of computing resources required for PepPIs prediction. According to comparison analysis, the accuracy reached 0.855 in TPepPro, achieving an 8.1% improvement compared to the second-best model TAGPPI. TPepPro achieved an AUC of 0.922, surpassing the second-best model TAGPPI with 0.844. Moreover, the newly developed TPepPro identify certain PepPIs that can be validated according to previous experimental evidence, thus indicating the efficiency of TPepPro to detect high potential PepPIs that would be helpful for amino acid drug applications.

AVAILABILITY AND IMPLEMENTATION

The source code of TPepPro is available at https://github.com/wanglabhku/TPepPro.

Predicting adaptive immune receptor specificities by machine learning is a data generation problem

Determining the specificity of adaptive immune receptors-B cell receptors (BCRs), their secreted form antibodies, and T cell receptors (TCRs)-is critical for understanding immune responses and advancing immunotherapy and drug discovery. Immune receptors exhibit extensive diversity in their variable domains, enabling them to interact with a plethora of antigens. Despite the significant progress made by AI tools such as AlphaFold in predicting protein structures, challenges remain in accurately modeling the structure and specificity of immune receptors, primarily due to the limited availability of high-quality crystal structures and the complexity of immune receptor-antigen interactions. In this perspective, we highlight recent advancements in sequence-based and structure-based data generation for immune receptors, which are crucial for training machine learning models that predict receptor specificity. We discuss the current bottlenecks and potential future directions in generating and utilizing high-dimensional datasets for predicting and designing the specificity of antibodies and TCRs.

Expanding the landscape of antibody discovery

Library:library screening technologies hold substantial promise for paired antibody:antigen discovery, but challenges have persisted. In this issue of Cell Reports Methods, Wagner et al. introduce a method that combines antibody-ribosome-mRNA complexes, antigen cell surface display, and single-cell RNA sequencing to successfully screen diverse antibody gene libraries against a library of viral receptor proteins.

Graph Neural Networks-Based Prediction of Drug Gene Interactions of RTK-VEGF4 Receptor Family in Periodontal Regeneration

Background

The RTK-VEGF4 receptor family, which includes VEGFR-1, VEGFR-2, and VEGFR-3, plays a crucial role in tissue regeneration by promoting angiogenesis, the formation of new blood vessels, and recruiting stem cells and immune cells. Machine learning, particularly graph neural networks (GNNs), has shown high accuracy in predicting these interactions. This study aims to predict drug-gene interactions of the RTK-VEGF4 receptor family in periodontal regeneration using graph neural networks.

Material and Methods

The study utilized a dataset comprising 19,154 drug-gene interactions to analyze the relationships between drugs and protein-coding genes. The dataset was split into training and testing sets, with 80% of the data used for training and 20% for testing. Cytoscape, an open-source software platform, was employed to visualize and analyze the drug-gene interaction network, and CytoHubba, a plugin, was used to identify highly connected nodes. Topological measures were applied to determine the influence and importance of each node. GNNs were used to manage the complex relationships and dependencies within the graphs.

Results

The drug-gene interaction network, comprising 815 nodes and 13,436 edges, was found to be complex and highly interconnected. It was divided into 11 components, displaying low density and heterogeneity, indicative of a sparse structure. The GNN model achieved 97% accuracy in predicting interaction types, including single protein interactions and protein complex groups.

Conclusions

The study demonstrates that graph neural networks outperform traditional machine learning methods in predicting drug-gene interactions within the RTK-VEGF protein family in periodontal regeneration, highlighting their potential in advancing therapeutic strategies and drug discovery. Key words:Graph neural networks; drug-gene interactions; RTK-VEGF4 protein family: periodontal regeneration.

Mass Spectrometry Probe Combined with Machine Learning to Capture the Relationship between Metabolites and Mitochondrial Complex Activity at the Whole-Cell Level

Mitochondrial complex activity controls a multitude of physiological processes by regulating the cellular metabolism. Current methods for evaluating mitochondrial complex activity mainly focus on single metabolic reactions within mitochondria. These methods often require fresh samples in large quantities for mitochondria purification or intact mitochondrial membranes for real-time monitoring. Confronting these limitations, we shifted the analytical perspective toward interactive metabolic networks at the whole-cell level to reflect mitochondrial complex activity. To this end, we compiled a panel of mitochondrial respiratory chain-mapped metabolites (MRCMs), whose perturbations theoretically provide an overall reflection on mitochondrial complex activity. By introducing N-dimethyl-p-phenylenediamine and N-methyl-p-phenylenediamine as a pair of mass spectrometry probes, an ultraperformance liquid chromatography-tandem mass spectrometry method with high sensitivity (LLOQ as low as 0.2 fmol) was developed to obtain accurate quantitative data of MRCMs. Machine learning was then combined to capture the relationship between MRCMs and mitochondrial complex activity. Using Complex I as a proof-of-concept, we identified NADH, alanine, and phosphoenolpyruvate as metabolites associated with Complex I activity based on the whole-cell level. The effectiveness of using their concentrations to reflect Complex I activity was further validated in external data sets. Hence, by capturing the relationship between metabolites and mitochondrial complex activity at the whole-cell level, this study explores a novel analytical paradigm for the interrogation of mitochondrial complex activity, offering a favorable complement to existing methods particularly when sample quantities, type, and treatment timeliness pose challenges. More importantly, it shifts the focus from individual metabolic reactions within mitochondria to a more comprehensive view of an interactive metabolic network, which should serve as a promising direction for future research into the functional architecture between mitochondrial complexes and metabolites.

Yeast-based screening platforms to understand and improve human health

Detailed molecular understanding of the human organism is essential to develop effective therapies. Saccharomyces cerevisiae has been used extensively for acquiring insights into important aspects of human health, such as studying genetics and cell-cell communication, elucidating protein-protein interaction (PPI) networks, and investigating human G protein-coupled receptor (hGPCR) signaling. We highlight recent advances and opportunities of yeast-based technologies for cost-efficient chemical library screening on hGPCRs, accelerated deciphering of PPI networks with mating-based screening and selection, and accurate cell-cell communication with human immune cells. Overall, yeast-based technologies constitute an important platform to support basic understanding and innovative applications towards improving human health.

Reactive Oxygen Species Mechanisms that Regulate Protein-Protein Interactions in Cancer

Reactive oxygen species (ROS) are produced during cellular metabolism and in response to environmental stress. While low levels of ROS play essential physiological roles, excess ROS can damage cellular components, leading to cell death or transformation. ROS can also regulate protein interactions in cancer cells, thereby affecting processes such as cell growth, migration, and angiogenesis. Dysregulated interactions occur via various mechanisms, including amino acid modifications, conformational changes, and alterations in complex stability. Understanding ROS-mediated changes in protein interactions is crucial for targeted cancer therapies. In this review, we examine the role that ROS mechanisms in regulating pathways through protein-protein interactions.

Addressing epistasis in the design of protein function

Mutations in protein active sites can dramatically improve function. The active site, however, is densely packed and extremely sensitive to mutations. Therefore, some mutations may only be tolerated in combination with others in a phenomenon known as epistasis. Epistasis reduces the likelihood of obtaining improved functional variants and dramatically slows natural and lab evolutionary processes. Research has shed light on the molecular origins of epistasis and its role in shaping evolutionary trajectories and outcomes. In addition, sequence- and AI-based strategies that infer epistatic relationships from mutational patterns in natural or experimental evolution data have been used to design functional protein variants. In recent years, combinations of such approaches and atomistic design calculations have successfully predicted highly functional combinatorial mutations in active sites. These were used to design thousands of functional active-site variants, demonstrating that, while our understanding of epistasis remains incomplete, some of the determinants that are critical for accurate design are now sufficiently understood. We conclude that the space of active-site variants that has been explored by evolution may be expanded dramatically to enhance natural activities or discover new ones. Furthermore, design opens the way to systematically exploring sequence and structure space and mutational impacts on function, deepening our understanding and control over protein activity.

Computational tools for plant genomics and breeding

Plant genomics and crop breeding are at the intersection of biotechnology and information technology. Driven by a combination of high-throughput sequencing, molecular biology and data science, great advances have been made in omics technologies at every step along the central dogma, especially in genome assembling, genome annotation, epigenomic profiling, and transcriptome profiling. These advances further revolutionized three directions of development. One is genetic dissection of complex traits in crops, along with genomic prediction and selection. The second is comparative genomics and evolution, which open up new opportunities to depict the evolutionary constraints of biological sequences for deleterious variant discovery. The third direction is the development of deep learning approaches for the rational design of biological sequences, especially proteins, for synthetic biology. All three directions of development serve as the foundation for a new era of crop breeding where agronomic traits are enhanced by genome design.

Assessing the functional roles of coevolving PHD finger residues

Although in silico folding based on coevolving residue constraints in the deep-learning era has transformed protein structure prediction, the contributions of coevolving residues to protein folding, stability, and other functions in physical contexts remain to be clarified and experimentally validated. Herein, the PHD finger module, a well-known histone reader with distinct subtypes containing subtype-specific coevolving residues, was used as a model to experimentally assess the contributions of coevolving residues and to clarify their specific roles. The results of the assessment, including proteolysis and thermal unfolding of wildtype and mutant proteins, suggested that coevolving residues have varying contributions, despite their large in silico constraints. Residue positions with large constraints were found to contribute to stability in one subtype but not others. Computational sequence design and generative model-based energy estimates of individual structures were also implemented to complement the experimental assessment. Sequence design and energy estimates distinguish coevolving residues that contribute to folding from those that do not. The results of proteolytic analysis of mutations at positions contributing to folding were consistent with those suggested by sequence design and energy estimation. Thus, we report a comprehensive assessment of the contributions of coevolving residues, as well as a strategy based on a combination of approaches that should enable detailed understanding of the residue contributions in other large protein families.

Best practices for machine learning in antibody discovery and development

In the past 40 years, therapeutic antibody discovery and development have advanced considerably, with machine learning (ML) offering a promising way to speed up the process by reducing costs and the number of experiments required. Recent progress in ML-guided antibody design and development (D&D) has been hindered by the diversity of data sets and evaluation methods, which makes it difficult to conduct comparisons and assess utility. Establishing standards and guidelines will be crucial for the wider adoption of ML and the advancement of the field. This perspective critically reviews current practices, highlights common pitfalls and proposes method development and evaluation guidelines for various ML-based techniques in therapeutic antibody D&D. Addressing challenges across the ML process, best practices are recommended for each stage to enhance reproducibility and progress.

A library-on-library screen reveals the breadth expansion landscape of a broadly neutralizing betacoronavirus antibody

Broadly neutralizing antibodies (bnAbs) typically evolve cross-reactivity breadth through acquiring somatic hypermutations. While evolution of breadth requires improvement of binding to multiple antigenic variants, most experimental evolution platforms select against only one antigenic variant at a time. In this study, a yeast display library-on-library approach was applied to delineate the affinity maturation of a betacoronavirus bnAb, S2P6, against 27 spike stem helix peptides in a single experiment. Our results revealed that the binding affinity landscape of S2P6 varies among different stem helix peptides. However, somatic hypermutations that confer general improvement in binding affinity across different stem helix peptides could also be identified. We further showed that a key somatic hypermutation for breadth expansion involves long-range interaction. Overall, our work not only provides a proof-of-concept for using a library-on-library approach to analyze the evolution of antibody breadth, but also has important implications for the development of broadly protective vaccines.

Advancing Antibody Engineering through Synthetic Evolution and Machine Learning

Abs are versatile molecules with the potential to achieve exceptional binding to target Ags, while also possessing biophysical properties suitable for therapeutic drug development. Protein display and directed evolution systems have transformed synthetic Ab discovery, engineering, and optimization, vastly expanding the number of Ab clones able to be experimentally screened for binding. Moreover, the burgeoning integration of high-throughput screening, deep sequencing, and machine learning has further augmented in vitro Ab optimization, promising to accelerate the design process and massively expand the Ab sequence space interrogated. In this Brief Review, we discuss the experimental and computational tools employed in synthetic Ab engineering and optimization. We also explore the therapeutic challenges posed by developing Abs for infectious diseases, and the prospects for leveraging machine learning-guided protein engineering to prospectively design Abs resistant to viral escape.