Rapid protein stability prediction using deep learning representations

Complex state transitions of the bacterial cell division protein FtsZ

The key bacterial cell division protein FtsZ can adopt multiple conformations, and prevailing models suggest that transitions of FtsZ subunits from the closed to open state are necessary for filament formation and stability. Using all-atom molecular dynamics simulations, we analyzed state transitions of Staphylococcus aureus FtsZ as a monomer, dimer, and hexamer. We found that monomers can adopt intermediate states but preferentially adopt a closed state that is robust to forced reopening. Dimer subunits transitioned between open and closed states, and dimers with both subunits in the closed state remained highly stable, suggesting that open-state conformations are not necessary for filament formation. Mg2+ strongly stabilized the conformation of GTP-bound subunits and the dimer filament interface. Our hexamer simulations indicate that the plus end subunit preferentially closes and that other subunits can transition between states without affecting inter-subunit stability. We found that rather than being correlated with subunit opening, inter-subunit stability was strongly correlated with catalytic site interactions. By leveraging deep-learning models, we identified key intrasubunit interactions governing state transitions. Our findings suggest a greater range of possible monomer and filament states than previously considered and offer new insights into the nuanced interplay between subunit states and the critical role of nucleotide hydrolysis and Mg2+ in FtsZ filament dynamics.

ASM variants in the spotlight: A structure-based atlas for unraveling pathogenic mechanisms in lysosomal acid sphingomyelinase

Lysosomal acid sphingomyelinase (ASM), a critical enzyme in lipid metabolism encoded by the SMPD1 gene, plays a crucial role in sphingomyelin hydrolysis in lysosomes. ASM deficiency leads to acid sphingomyelinase deficiency, a rare genetic disorder with diverse clinical manifestations, and the protein can be found mutated in other diseases. We employed a structure-based framework to comprehensively understand the functional implications of ASM variants, integrating pathogenicity predictions with molecular insights derived from a molecular dynamics simulation in a lysosomal membrane environment. Our analysis, encompassing over 400 variants, establishes a structural atlas of missense variants of lysosomal ASM, associating mechanistic indicators with pathogenic potential. Our study highlights variants that influence structural stability or exert local and long-range effects at functional sites. To validate our predictions, we compared them to available experimental data on residual catalytic activity in 135 ASM variants. Notably, our findings also suggest applications of the resulting data for identifying cases suited for enzyme replacement therapy. This comprehensive approach enhances the understanding of ASM variants and provides valuable insights for potential therapeutic interventions.

An end-to-end framework for the prediction of protein structure and fitness from single sequence

Significant research progress has been made in the field of protein structure and fitness prediction. Particularly, single-sequence-based structure prediction methods like ESMFold and OmegaFold achieve a balance between inference speed and prediction accuracy, showing promise for many downstream prediction tasks. Here, we propose SPIRED, a single-sequence-based structure prediction model that exhibits comparable performance to the state-of-the-art methods but with approximately 5-fold acceleration in inference and at least one order of magnitude reduction in training consumption. By integrating SPIRED with downstream neural networks, we compose an end-to-end framework named SPIRED-Fitness for the rapid prediction of both protein structure and fitness from single sequence with satisfactory accuracy. Moreover, SPIRED-Stab, the derivative of SPIRED-Fitness, achieves state-of-the-art performance in predicting the mutational effects on protein stability.

Deciphering GB1's Single Mutational Landscape: Insights from MuMi Analysis

Mutational changes that affect the binding of the C2 fragment of Streptococcal protein G (GB1) to the Fc domain of human IgG (IgG-Fc) have been extensively studied using deep mutational scanning (DMS), and the binding affinity of all single mutations has been measured experimentally in the literature. To investigate the underlying molecular basis, we perform in silico mutational scanning for all possible single mutations, along with 2 μs-long molecular dynamics (WT-MD) of the wild-type (WT) GB1 in both unbound and IgG-Fc bound forms. We compute the hydrogen bonds between GB1 and IgG-Fc in WT-MD to identify the dominant hydrogen bonds for binding, which we then assess in conformations produced by Mutation and Minimization (MuMi) to explain the fitness landscape of GB1 and IgG-Fc binding. Furthermore, we analyze MuMi and WT-MD to investigate the dynamics of binding, focusing on the relative solvent accessibility of residues and the probability of residues being located at the binding interface. With these analyses, we explain the interactions between GB1 and IgG-Fc and display the structural features of binding. In sum, our findings highlight the potential of MuMi as a reliable and computationally efficient tool for predicting protein fitness landscapes, offering significant advantages over traditional methods. The methodologies and results presented in this study pave the way for improved predictive accuracy in protein stability and interaction studies, which are crucial for advancements in drug design and synthetic biology.

Improved Prediction of Stabilizing Mutations in Proteins by Incorporation of Mutational Effects on Ligand Binding

While many computational methods accurately predict destabilizing mutations, identifying stabilizing mutations has remained a challenge, because of their relative rarity. We tested ΔΔG0 predictions from computational predictors such as Rosetta, ThermoMPNN, RaSP, and DeepDDG, using 82 mutants of the bacterial toxin CcdB as a test case. On this dataset, the best computational predictor is ThermoMPNN, which identifies stabilizing mutations with a precision of 68%. However, the average increase in Tm for these predicted mutations was only 1°C for CcdB, and predictions were poorer for a more challenging target, influenza neuraminidase. Using data from multiple previously described yeast surface display libraries and in vitro thermal stability measurements, we trained logistic regression models to identify stabilizing mutations with a precision of 90% and an average increase in Tm of 3°C for CcdB. When such libraries contain a population of mutants with significantly enhanced binding relative to the corresponding wild type, there is no benefit in using computational predictors. It is then possible to predict stabilizing mutations without any training, simply by examining the distribution of mutational binding scores. This avoids laborious steps of in vitro expression, purification, and stability characterization. When this is not the case, combining data from computational predictors with high-throughput experimental binding data enhances the prediction of stabilizing mutations. However, this requires training on stability data measured in vitro with known stabilized mutants. It is thus feasible to predict stabilizing mutations rapidly and accurately for any system of interest that can be subjected to a binding selection or screen.

Integrating Large-Scale Protein Structure Prediction into Human Genetics Research

The last five years have seen impressive progress in deep learning models applied to protein research. Most notably, sequence-based structure predictions have seen transformative gains in the form of AlphaFold2 and related approaches. Millions of missense protein variants in the human population lack annotations, and these computational methods are a valuable means to prioritize variants for further analysis. Here, we review the recent progress in deep learning models applied to the prediction of protein structure and protein variants, with particular emphasis on their implications for human genetics and health. Improved prediction of protein structures facilitates annotations of the impact of variants on protein stability, protein-protein interaction interfaces, and small-molecule binding pockets. Moreover, it contributes to the study of host-pathogen interactions and the characterization of protein function. As genome sequencing in large cohorts becomes increasingly prevalent, we believe that better integration of state-of-the-art protein informatics technologies into human genetics research is of paramount importance.

Stability Oracle: a structure-based graph-transformer framework for identifying stabilizing mutations

Engineering stabilized proteins is a fundamental challenge in the development of industrial and pharmaceutical biotechnologies. We present Stability Oracle: a structure-based graph-transformer framework that achieves SOTA performance on accurately identifying thermodynamically stabilizing mutations. Our framework introduces several innovations to overcome well-known challenges in data scarcity and bias, generalization, and computation time, such as: Thermodynamic Permutations for data augmentation, structural amino acid embeddings to model a mutation with a single structure, a protein structure-specific attention-bias mechanism that makes transformers a viable alternative to graph neural networks. We provide training/test splits that mitigate data leakage and ensure proper model evaluation. Furthermore, to examine our data engineering contributions, we fine-tune ESM2 representations (Prostata-IFML) and achieve SOTA for sequence-based models. Notably, Stability Oracle outperforms Prostata-IFML even though it was pretrained on 2000X less proteins and has 548X less parameters. Our framework establishes a path for fine-tuning structure-based transformers to virtually any phenotype, a necessary task for accelerating the development of protein-based biotechnologies.

Enhancing predictions of protein stability changes induced by single mutations using MSA-based Language Models

MOTIVATION

Protein Language Models offer a new perspective for addressing challenges in structural biology, while relying solely on sequence information. Recent studies have investigated their effectiveness in forecasting shifts in thermodynamic stability caused by single amino acid mutations, a task known for its complexity due to the sparse availability of data, constrained by experimental limitations. To tackle this problem, we introduce two key novelties: leveraging a Protein Language Model that incorporates Multiple Sequence Alignments to capture evolutionary information, and using a recently released mega-scale dataset with rigorous data pre-processing to mitigate overfitting.

RESULTS

We ensure comprehensive comparisons by fine-tuning various pre-trained models, taking advantage of analyses such as ablation studies and baselines evaluation. Our methodology introduces a stringent policy to reduce the widespread issue of data leakage, rigorously removing sequences from the training set when they exhibit significant similarity with the test set. The MSA Transformer emerges as the most accurate among the models under investigation, given its capability to leverage co-evolution signals encoded in aligned homologous sequences. Moreover, the optimized MSA Transformer outperforms existing methods and exhibits enhanced generalization power, leading to a notable improvement in predicting changes in protein stability resulting from point mutations.

AVAILABILITY AND IMPLEMENTATION

Code and data at https://github.com/RitAreaSciencePark/PLM4Muts.

SUPPLEMENTARY INFORMATION

Supplementary Information is available at Bioinformatics online.

Novel pathogenic variants of DNAH5 associated with clinical and genetic spectra of primary ciliary dyskinesia in an Arab population

Introduction: Primary ciliary dyskinesia (PCD) is caused by the dysfunction of motile cilia resulting in insufficient mucociliary clearance of the lungs. This study aimed to map novel PCD variants and determine their pathogenicity in PCD patients in Kuwait. Methods: Herein, we present five PCD individuals belonging to a cohort of 105 PCD individuals recruited from different hospitals in Kuwait. Genomic DNAs from the family members were analysed to screen for pathogenic PCD variants. Transmission electron microscopy (TEM) and immunofluorescence (IF) analyses were performed on the nasal biopsies to detect specific structural abnormalities within the ciliated cells. Results: Genetic screening and functional analyses confirmed that the five PCD individuals carried novel pathogenic variants of DNAH5 causing PCD in three Arabic families. Of these, one multiplex family with two affected individuals showed two novel homozygous missense variants in DNAH5 causing PCD with situs inversus; another multiplex family with two affected individuals showed two newly identified compound heterozygous variants in DNAH5 causing PCD with situs solitus. In addition, novel heterozygous variants were identified in a child with PCD and situs solitus from a singleton family with unrelated parents. TEM analysis demonstrated the lack of outer dynein arms (ODAs) in all analysed samples, and IF analysis confirmed the absence of the dynein arm component of DNAH5 from the ciliary axoneme. Conclusion: The newly identified pathogenic variants of DNAH5 are associated with PCD as well as variable pulmonary clinical manifestations in Arabic families.

HERMES: Holographic Equivariant neuRal network model for Mutational Effect and Stability prediction

Predicting the stability and fitness effects of amino acid mutations in proteins is a cornerstone of biological discovery and engineering. Various experimental techniques have been developed to measure mutational effects, providing us with extensive datasets across a diverse range of proteins. By training on these data, traditional computational modeling and more recent machine learning approaches have advanced significantly in predicting mutational effects. Here, we introduce HERMES, a 3D rotationally equivariant structure-based neural network model for mutational effect and stability prediction. Pre-trained to predict amino acid propensity from its surrounding 3D structure, HERMES can be fine-tuned for mutational effects using our open-source code. We present a suite of HERMES models, pre-trained with different strategies, and fine-tuned to predict the stability effect of mutations. Benchmarking against other models shows that HERMES often outperforms or matches their performance in predicting mutational effect on stability, binding, and fitness. HERMES offers versatile tools for evaluating mutational effects and can be fine-tuned for specific predictive objectives.

MutationExplorer: a webserver for mutation of proteins and 3D visualization of energetic impacts

The possible effects of mutations on stability and function of a protein can only be understood in the context of protein 3D structure. The MutationExplorer webserver maps sequence changes onto protein structures and allows users to study variation by inputting sequence changes. As the user enters variants, the 3D model evolves, and estimated changes in energy are highlighted. In addition to a basic per-residue input format, MutationExplorer can also upload an entire replacement sequence. Previously the purview of desktop applications, such an upload can back-mutate PDB structures to wildtype sequence in a single step. Another supported variation source is human single nucelotide polymorphisms (SNPs), genomic coordinates input in VCF format. Structures are flexibly colorable, not only by energetic differences, but also by hydrophobicity, sequence conservation, or other biochemical profiling. Coloring by interface score reveals mutation impacts on binding surfaces. MutationExplorer strives for efficiency in user experience. For example, we have prepared 45 000 PDB depositions for instant retrieval and initial display. All modeling steps are performed by Rosetta. Visualizations leverage MDsrv/Mol*. MutationExplorer is available at: http://proteinformatics.org/mutation_explorer/.

Protein stability prediction by fine-tuning a protein language model on a mega-scale dataset

Protein stability plays a crucial role in a variety of applications, such as food processing, therapeutics, and the identification of pathogenic mutations. Engineering campaigns commonly seek to improve protein stability, and there is a strong interest in streamlining these processes to enable rapid optimization of highly stabilized proteins with fewer iterations. In this work, we explore utilizing a mega-scale dataset to develop a protein language model optimized for stability prediction. ESMtherm is trained on the folding stability of 528k natural and de novo sequences derived from 461 protein domains and can accommodate deletions, insertions, and multiple-point mutations. We show that a protein language model can be fine-tuned to predict folding stability. ESMtherm performs reasonably on small protein domains and generalizes to sequences distal from the training set. Lastly, we discuss our model's limitations compared to other state-of-the-art methods in generalizing to larger protein scaffolds. Our results highlight the need for large-scale stability measurements on a diverse dataset that mirrors the distribution of sequence lengths commonly observed in nature.

Destabilization and Degradation of a Disease-Linked PGM1 Protein Variant

PGM1-linked congenital disorder of glycosylation (PGM1-CDG) is an autosomal recessive disease characterized by several phenotypes, some of which are life-threatening. Research focusing on the disease-related variants of the α-D-phosphoglucomutase 1 (PGM1) protein has shown that several are insoluble in vitro and expressed at low levels in patient fibroblasts. Due to these observations, we hypothesized that some disease-linked PGM1 protein variants are structurally destabilized and subject to protein quality control (PQC) and rapid intracellular degradation. Employing yeast-based assays, we show that a disease-associated human variant, PGM1 L516P, is insoluble, inactive, and highly susceptible to ubiquitylation and rapid degradation by the proteasome. In addition, we show that PGM1 L516P forms aggregates in S. cerevisiae and that both the aggregation pattern and the abundance of PGM1 L516P are chaperone-dependent. Finally, using computational methods, we perform saturation mutagenesis to assess the impact of all possible single residue substitutions in the PGM1 protein. These analyses identify numerous missense variants with predicted detrimental effects on protein function and stability. We suggest that many disease-linked PGM1 variants are subject to PQC-linked degradation and that our in silico site-saturated data set may assist in the mechanistic interpretation of PGM1 variants.

Review of predicting protein stability changes upon variations

Forecasting alterations in protein stability caused by variations holds immense importance. Improving the thermal stability of proteins is important for biomedical and industrial applications. This review discusses the latest methods for predicting the effects of mutations on protein stability, databases containing protein mutations and thermodynamic parameters, and experimental techniques for efficiently assessing protein stability in high-throughput settings. Various publicly available databases for protein stability prediction are introduced. Furthermore, state-of-the-art computational approaches for anticipating protein stability changes due to variants are reviewed. Each method's types of features, base algorithm, and prediction results are also detailed. Additionally, some experimental approaches for verifying the prediction results of computational methods are introduced. Finally, the review summarizes the progress and challenges of protein stability prediction and discusses potential models for future research directions.

Predicted mechanistic impacts of human protein missense variants

Genome sequencing efforts have led to the discovery of tens of millions of protein missense variants found in the human population with the majority of these having no annotated role and some likely contributing to trait variation and disease. Sequence-based artificial intelligence approaches have become highly accurate at predicting variants that are detrimental to the function of proteins but they do not inform on mechanisms of disruption. Here we combined sequence and structure-based methods to perform proteome-wide prediction of deleterious variants with information on their impact on protein stability, protein-protein interactions and small-molecule binding pockets. AlphaFold2 structures were used to predict approximately 100,000 small-molecule binding pockets and stability changes for over 200 million variants. To inform on protein-protein interfaces we used AlphaFold2 to predict structures for nearly 500,000 protein complexes. We illustrate the value of mechanism-aware variant effect predictions to study the relation between protein stability and abundance and the structural properties of interfaces underlying trans protein quantitative trait loci (pQTLs). We characterised the distribution of mechanistic impacts of protein variants found in patients and experimentally studied example disease linked variants in FGFR1.

Deep mutational scanning reveals a correlation between degradation and toxicity of thousands of aspartoacylase variants

Unstable proteins are prone to form non-native interactions with other proteins and thereby may become toxic. To mitigate this, destabilized proteins are targeted by the protein quality control network. Here we present systematic studies of the cytosolic aspartoacylase, ASPA, where variants are linked to Canavan disease, a lethal neurological disorder. We determine the abundance of 6152 of the 6260 ( ~ 98%) possible single amino acid substitutions and nonsense ASPA variants in human cells. Most low abundance variants are degraded through the ubiquitin-proteasome pathway and become toxic upon prolonged expression. The data correlates with predicted changes in thermodynamic stability, evolutionary conservation, and separate disease-linked variants from benign variants. Mapping of degradation signals (degrons) shows that these are often buried and the C-terminal region functions as a degron. The data can be used to interpret Canavan disease variants and provide insight into the relationship between protein stability, degradation and cell fitness.

Genotype sampling for deep-learning assisted experimental mapping of a combinatorially complete fitness landscape

MOTIVATION

Experimental characterization of fitness landscapes, which map genotypes onto fitness, is important for both evolutionary biology and protein engineering. It faces a fundamental obstacle in the astronomical number of genotypes whose fitness needs to be measured for any one protein. Deep learning may help to predict the fitness of many genotypes from a smaller neural network training sample of genotypes with experimentally measured fitness. Here I use a recently published experimentally mapped fitness landscape of more than 260 000 protein genotypes to ask how such sampling is best performed.

RESULTS

I show that multilayer perceptrons, recurrent neural networks, convolutional networks, and transformers, can explain more than 90% of fitness variance in the data. In addition, 90% of this performance is reached with a training sample comprising merely ≈103 sequences. Generalization to unseen test data is best when training data is sampled randomly and uniformly, or sampled to minimize the number of synonymous sequences. In contrast, sampling to maximize sequence diversity or codon usage bias reduces performance substantially. These observations hold for more than one network architecture. Simple sampling strategies may perform best when training deep learning neural networks to map fitness landscapes from experimental data.

AVAILABILITY AND IMPLEMENTATION

The fitness landscape data analyzed here is publicly available as described previously (Papkou et al. 2023). All code used to analyze this landscape is publicly available at https://github.com/andreas-wagner-uzh/fitness_landscape_sampling.

Prediction of pyrazinamide resistance in Mycobacterium tuberculosis using structure-based machine-learning approaches

Background

Pyrazinamide is one of four first-line antibiotics used to treat tuberculosis; however, antibiotic susceptibility testing for pyrazinamide is challenging. Resistance to pyrazinamide is primarily driven by genetic variation in pncA, encoding an enzyme that converts pyrazinamide into its active form.

Methods

We curated a dataset of 664 non-redundant, missense amino acid mutations in PncA with associated high-confidence phenotypes from published studies and then trained three different machine-learning models to predict pyrazinamide resistance. All models had access to a range of protein structural-, chemical- and sequence-based features.

Results

The best model, a gradient-boosted decision tree, achieved a sensitivity of 80.2% and a specificity of 76.9% on the hold-out test dataset. The clinical performance of the models was then estimated by predicting the binary pyrazinamide resistance phenotype of 4027 samples harbouring 367 unique missense mutations in pncA derived from 24 231 clinical isolates.

Conclusions

This work demonstrates how machine learning can enhance the sensitivity/specificity of pyrazinamide resistance prediction in genetics-based clinical microbiology workflows, highlights novel mutations for future biochemical investigation, and is a proof of concept for using this approach in other drugs.

Biophysics-based protein language models for protein engineering

Protein language models trained on evolutionary data have emerged as powerful tools for predictive problems involving protein sequence, structure, and function. However, these models overlook decades of research into biophysical factors governing protein function. We propose Mutational Effect Transfer Learning (METL), a protein language model framework that unites advanced machine learning and biophysical modeling. Using the METL framework, we pretrain transformer-based neural networks on biophysical simulation data to capture fundamental relationships between protein sequence, structure, and energetics. We finetune METL on experimental sequence-function data to harness these biophysical signals and apply them when predicting protein properties like thermostability, catalytic activity, and fluorescence. METL excels in challenging protein engineering tasks like generalizing from small training sets and position extrapolation, although existing methods that train on evolutionary signals remain powerful for many types of experimental assays. We demonstrate METL's ability to design functional green fluorescent protein variants when trained on only 64 examples, showcasing the potential of biophysics-based protein language models for protein engineering.

MUTATIONEXPLORER- A WEBSERVER FOR MUTATION OF PROTEINS AND 3D VISUALIZATION OF ENERGETIC IMPACTS

The possible effects of mutations on stability and function of a protein can only be understood in the context of protein 3D structure. The MutationExplorer webserver maps sequence changes onto protein structures and allows users to study variation by inputting sequence changes. As the user enters variants, the 3D model evolves, and estimated changes in energy are highlighted. In addition to a basic per-residue input format, MutationExplorer can also upload an entire replacement sequence. Previously the purview of desktop applications, such an upload can back-mutate PDB structures to wildtype sequence in a single step. Another supported variation source is human single nucelotide polymorphisms (SNPs), genomic coordinates input in VCF format.

A mutational atlas for Parkin proteostasis

Proteostasis can be disturbed by mutations affecting folding and stability of the encoded protein. An example is the ubiquitin ligase Parkin, where gene variants result in autosomal recessive Parkinsonism. To uncover the pathological mechanism and provide comprehensive genotype-phenotype information, variant abundance by massively parallel sequencing (VAMP-seq) is leveraged to quantify the abundance of Parkin variants in cultured human cells. The resulting mutational map, covering 9219 out of the 9300 possible single-site amino acid substitutions and nonsense Parkin variants, shows that most low abundance variants are proteasome targets and are located within the structured domains of the protein. Half of the known disease-linked variants are found at low abundance. Systematic mapping of degradation signals (degrons) reveals an exposed degron region proximal to the so-called "activation element". This work provides examples of how missense variants may cause degradation either via destabilization of the native protein, or by introducing local signals for degradation.

Learning the shape of protein microenvironments with a holographic convolutional neural network

Proteins play a central role in biology from immune recognition to brain activity. While major advances in machine learning have improved our ability to predict protein structure from sequence, determining protein function from its sequence or structure remains a major challenge. Here, we introduce holographic convolutional neural network (H-CNN) for proteins, which is a physically motivated machine learning approach to model amino acid preferences in protein structures. H-CNN reflects physical interactions in a protein structure and recapitulates the functional information stored in evolutionary data. H-CNN accurately predicts the impact of mutations on protein stability and binding of protein complexes. Our interpretable computational model for protein structure-function maps could guide design of novel proteins with desired function.

Transfer learning to leverage larger datasets for improved prediction of protein stability changes

Amino acid mutations that lower a protein's thermodynamic stability are implicated in numerous diseases, and engineered proteins with enhanced stability can be important in research and medicine. Computational methods for predicting how mutations perturb protein stability are, therefore, of great interest. Despite recent advancements in protein design using deep learning, in silico prediction of stability changes has remained challenging, in part due to a lack of large, high-quality training datasets for model development. Here, we describe ThermoMPNN, a deep neural network trained to predict stability changes for protein point mutations given an initial structure. In doing so, we demonstrate the utility of a recently released megascale stability dataset for training a robust stability model. We also employ transfer learning to leverage a second, larger dataset by using learned features extracted from ProteinMPNN, a deep neural network trained to predict a protein's amino acid sequence given its three-dimensional structure. We show that our method achieves state-of-the-art performance on established benchmark datasets using a lightweight model architecture that allows for rapid, scalable predictions. Finally, we make ThermoMPNN readily available as a tool for stability prediction and design.

Valorization of single-carbon chemicals by using carboligases as key enzymes

Single-carbon (C1) biorefinery plays a key role in the consumption of global greenhouse gases and a circular carbon economy. Thereby, we have focused on the valorization of C1 compounds (e.g. methanol, formaldehyde, and formate) into multicarbon products, including bioplastic monomers, glycolate, and ethylene glycol. For instance, methanol, derived from the oxidation of CH4, can be converted into glycolate, ethylene glycol, or erythrulose via formaldehyde and glycolaldehyde, employing C1 and/or C2 carboligases as essential enzymes. Escherichia coli was engineered to convert formate, produced from CO via CO2 or from CO2 directly, into glycolate. Recent progress in the design of biotransformation pathways, enzyme discovery, and engineering, as well as whole-cell biocatalyst engineering for C1 biorefinery, was addressed in this review.

Machine learning for functional protein design

Recent breakthroughs in AI coupled with the rapid accumulation of protein sequence and structure data have radically transformed computational protein design. New methods promise to escape the constraints of natural and laboratory evolution, accelerating the generation of proteins for applications in biotechnology and medicine. To make sense of the exploding diversity of machine learning approaches, we introduce a unifying framework that classifies models on the basis of their use of three core data modalities: sequences, structures and functional labels. We discuss the new capabilities and outstanding challenges for the practical design of enzymes, antibodies, vaccines, nanomachines and more. We then highlight trends shaping the future of this field, from large-scale assays to more robust benchmarks, multimodal foundation models, enhanced sampling strategies and laboratory automation.

Assessing computational tools for predicting protein stability changes upon missense mutations using a new dataset

Insight into how mutations affect protein stability is crucial for protein engineering, understanding genetic diseases, and exploring protein evolution. Numerous computational methods have been developed to predict the impact of amino acid substitutions on protein stability. Nevertheless, comparing these methods poses challenges due to variations in their training data. Moreover, it is observed that they tend to perform better at predicting destabilizing mutations than stabilizing ones. Here, we meticulously compiled a new dataset from three recently published databases: ThermoMutDB, FireProtDB, and ProThermDB. This dataset, which does not overlap with the well-established S2648 dataset, consists of 4038 single-point mutations, including over 1000 stabilizing mutations. We assessed these mutations using 27 computational methods, including the latest ones utilizing mega-scale stability datasets and transfer learning. We excluded entries with overlap or similarity to training datasets to ensure fairness. Pearson correlation coefficients for the tested tools ranged from 0.20 to 0.53 on unseen data, and none of the methods could accurately predict stabilizing mutations, even those performing well in anti-symmetric property analysis. While most methods present consistent trends for predicting destabilizing mutations across various properties such as solvent exposure and secondary conformation, stabilizing mutations do not exhibit a clear pattern. Our study also suggests that solely addressing training dataset bias may not significantly enhance accuracy of predicting stabilizing mutations. These findings emphasize the importance of developing precise predictive methods for stabilizing mutations.

FireProt 2.0: web-based platform for the fully automated design of thermostable proteins

Thermostable proteins find their use in numerous biomedical and biotechnological applications. However, the computational design of stable proteins often results in single-point mutations with a limited effect on protein stability. However, the construction of stable multiple-point mutants can prove difficult due to the possibility of antagonistic effects between individual mutations. FireProt protocol enables the automated computational design of highly stable multiple-point mutants. FireProt 2.0 builds on top of the previously published FireProt web, retaining the original functionality and expanding it with several new stabilization strategies. FireProt 2.0 integrates the AlphaFold database and the homology modeling for structure prediction, enabling calculations starting from a sequence. Multiple-point designs are constructed using the Bron-Kerbosch algorithm minimizing the antagonistic effect between the individual mutations. Users can newly limit the FireProt calculation to a set of user-defined mutations, run a saturation mutagenesis of the whole protein or select rigidifying mutations based on B-factors. Evolution-based back-to-consensus strategy is complemented by ancestral sequence reconstruction. FireProt 2.0 is significantly faster and a reworked graphical user interface broadens the tool's availability even to users with older hardware. FireProt 2.0 is freely available at http://loschmidt.chemi.muni.cz/fireprotweb.

Transfer learning to leverage larger datasets for improved prediction of protein stability changes

Amino acid mutations that lower a protein's thermodynamic stability are implicated in numerous diseases, and engineered proteins with enhanced stability are important in research and medicine. Computational methods for predicting how mutations perturb protein stability are therefore of great interest. Despite recent advancements in protein design using deep learning, in silico prediction of stability changes has remained challenging, in part due to a lack of large, high-quality training datasets for model development. Here we introduce ThermoMPNN, a deep neural network trained to predict stability changes for protein point mutations given an initial structure. In doing so, we demonstrate the utility of a newly released mega-scale stability dataset for training a robust stability model. We also employ transfer learning to leverage a second, larger dataset by using learned features extracted from a deep neural network trained to predict a protein's amino acid sequence given its three-dimensional structure. We show that our method achieves competitive performance on established benchmark datasets using a lightweight model architecture that allows for rapid, scalable predictions. Finally, we make ThermoMPNN readily available as a tool for stability prediction and design.

Discovering functionally important sites in proteins

Proteins play important roles in biology, biotechnology and pharmacology, and missense variants are a common cause of disease. Discovering functionally important sites in proteins is a central but difficult problem because of the lack of large, systematic data sets. Sequence conservation can highlight residues that are functionally important but is often convoluted with a signal for preserving structural stability. We here present a machine learning method to predict functional sites by combining statistical models for protein sequences with biophysical models of stability. We train the model using multiplexed experimental data on variant effects and validate it broadly. We show how the model can be used to discover active sites, as well as regulatory and binding sites. We illustrate the utility of the model by prospective prediction and subsequent experimental validation on the functional consequences of missense variants in HPRT1 which may cause Lesch-Nyhan syndrome, and pinpoint the molecular mechanisms by which they cause disease.

Correspondence between functional scores from deep mutational scans and predicted effects on protein stability

Many methodologically diverse computational methods have been applied to the growing challenge of predicting and interpreting the effects of protein variants. As many pathogenic mutations have a perturbing effect on protein stability or intermolecular interactions, one highly interpretable approach is to use protein structural information to model the physical impacts of variants and predict their likely effects on protein stability and interactions. Previous efforts have assessed the accuracy of stability predictors in reproducing thermodynamically accurate values and evaluated their ability to distinguish between known pathogenic and benign mutations. Here, we take an alternate approach, and explore how well stability predictor scores correlate with functional impacts derived from deep mutational scanning (DMS) experiments. In this work, we compare the predictions of 9 protein stability-based tools against mutant protein fitness values from 49 independent DMS datasets, covering 170,940 unique single amino acid variants. We find that FoldX and Rosetta show the strongest correlations with DMS-based functional scores, similar to their previous top performance in distinguishing between pathogenic and benign variants. For both methods, performance is considerably improved when considering intermolecular interactions from protein complex structures, when available. Furthermore, using these two predictors, we derive a "Foldetta" consensus score, which improves upon the performance of both, and manages to match dedicated variant effect predictors in reflecting variant functional impacts. Finally, we also highlight that predicted stability effects show consistently higher correlations with certain DMS experimental phenotypes, particularly those based upon protein abundance, and, in certain cases, can significantly outcompete sequence-based variant effect prediction methodologies for predicting functional scores from DMS experiments.

Rapid protein stability prediction using deep learning representations

Predicting the thermodynamic stability of proteins is a common and widely used step in protein engineering, and when elucidating the molecular mechanisms behind evolution and disease. Here, we present RaSP, a method for making rapid and accurate predictions of changes in protein stability by leveraging deep learning representations. RaSP performs on-par with biophysics-based methods and enables saturation mutagenesis stability predictions in less than a second per residue. We use RaSP to calculate ∼ 230 million stability changes for nearly all single amino acid changes in the human proteome, and examine variants observed in the human population. We find that variants that are common in the population are substantially depleted for severe destabilization, and that there are substantial differences between benign and pathogenic variants, highlighting the role of protein stability in genetic diseases. RaSP is freely available-including via a Web interface-and enables large-scale analyses of stability in experimental and predicted protein structures.