Scaffolding protein functional sites using deep learning

A Transformer-Based Ensemble Framework for the Prediction of Protein-Protein Interaction Sites

The identification of protein-protein interaction (PPI) sites is essential in the research of protein function and the discovery of new drugs. So far, a variety of computational tools based on machine learning have been developed to accelerate the identification of PPI sites. However, existing methods suffer from the low predictive accuracy or the limited scope of application. Specifically, some methods learned only global or local sequential features, leading to low predictive accuracy, while others achieved improved performance by extracting residue interactions from structures but were limited in their application scope for the serious dependence on precise structure information. There is an urgent need to develop a method that integrates comprehensive information to realize proteome-wide accurate profiling of PPI sites. Herein, a novel ensemble framework for PPI sites prediction, EnsemPPIS, was therefore proposed based on transformer and gated convolutional networks. EnsemPPIS can effectively capture not only global and local patterns but also residue interactions. Specifically, EnsemPPIS was unique in (a) extracting residue interactions from protein sequences with transformer and (b) further integrating global and local sequential features with the ensemble learning strategy. Compared with various existing methods, EnsemPPIS exhibited either superior performance or broader applicability on multiple PPI sites prediction tasks. Moreover, pattern analysis based on the interpretability of EnsemPPIS demonstrated that EnsemPPIS was fully capable of learning residue interactions within the local structure of PPI sites using only sequence information. The web server of EnsemPPIS is freely available at http://idrblab.org/ensemppis.

Searching similar local 3D micro-environments in protein structure databases with MicroMiner

The available protein structure data are rapidly increasing. Within these structures, numerous local structural sites depict the details characterizing structure and function. However, searching and analyzing these sites extensively and at scale poses a challenge. We present a new method to search local sites in protein structure databases using residue-defined local 3D micro-environments. We implemented the method in a new tool called MicroMiner and demonstrate the capabilities of residue micro-environment search on the example of structural mutation analysis. Usually, experimental structures for both the wild-type and the mutant are unavailable for comparison. With MicroMiner, we extracted $>255 \times 10^{6}$ amino acid pairs in protein structures from the PDB, exemplifying single mutations' local structural changes for single chains and $>45 \times 10^{6}$ pairs for protein-protein interfaces. We further annotate existing data sets of experimentally measured mutation effects, like $\Delta \Delta G$ measurements, with the extracted structure pairs to combine the mutation effect measurement with the structural change upon mutation. In addition, we show how MicroMiner can bridge the gap between mutation analysis and structure-based drug design tools. MicroMiner is available as a command line tool and interactively on the https://proteins.plus/ webserver.

Computational design of sequence-specific DNA-binding proteins

Sequence-specific DNA-binding proteins (DBPs) play critical roles in biology and biotechnology, and there has been considerable interest in the engineering of DBPs with new or altered specificities for genome editing and other applications. While there has been some success in reprogramming naturally occurring DBPs using selection methods, the computational design of new DBPs that recognize arbitrary target sites remains an outstanding challenge. We describe a computational method for the design of small DBPs that recognize specific target sequences through interactions with bases in the major groove, and employ this method in conjunction with experimental screening to generate binders for 5 distinct DNA targets. These binders exhibit specificity closely matching the computational models for the target DNA sequences at as many as 6 base positions and affinities as low as 30-100 nM. The crystal structure of a designed DBP-target site complex is in close agreement with the design model, highlighting the accuracy of the design method. The designed DBPs function in both Escherichia coli and mammalian cells to repress and activate transcription of neighboring genes. Our method is a substantial step towards a general route to small and hence readily deliverable sequence-specific DBPs for gene regulation and editing.

Designed Endocytosis-Triggering Proteins mediate Targeted Degradation

Endocytosis and lysosomal trafficking of cell surface receptors can be triggered by interaction with endogenous ligands. Therapeutic approaches such as LYTAC1,2 and KineTAC3, have taken advantage of this to target specific proteins for degradation by fusing modified native ligands to target binding proteins. While powerful, these approaches can be limited by possible competition with the endogenous ligand(s), the requirement in some cases for chemical modification that limits genetic encodability and can complicate manufacturing, and more generally, there may not be natural ligands which stimulate endocytosis through a given receptor. Here we describe general protein design approaches for designing endocytosis triggering binding proteins (EndoTags) that overcome these challenges. We present EndoTags for the IGF-2R, ASGPR, Sortillin, and Transferrin receptors, and show that fusing these tags to proteins which bind to soluble or transmembrane protein leads to lysosomal trafficking and target degradation; as these receptors have different tissue distributions, the different EndoTags could enable targeting of degradation to different tissues. The modularity and genetic encodability of EndoTags enables AND gate control for higher specificity targeted degradation, and the localized secretion of degraders from engineered cells. The tunability and modularity of our genetically encodable EndoTags should contribute to deciphering the relationship between receptor engagement and cellular trafficking, and they have considerable therapeutic potential as targeted degradation inducers, signaling activators for endocytosis-dependent pathways, and cellular uptake inducers for targeted antibody drug and RNA conjugates.

A probabilistic view of protein stability, conformational specificity, and design

Various approaches have used neural networks as probabilistic models for the design of protein sequences. These "inverse folding" models employ different objective functions, which come with trade-offs that have not been assessed in detail before. This study introduces probabilistic definitions of protein stability and conformational specificity and demonstrates the relationship between these chemical properties and the [Formula: see text] Boltzmann probability objective. This links the Boltzmann probability objective function to experimentally verifiable outcomes. We propose a novel sequence decoding algorithm, referred to as "BayesDesign", that leverages Bayes' Rule to maximize the [Formula: see text] objective instead of the [Formula: see text] objective common in inverse folding models. The efficacy of BayesDesign is evaluated in the context of two protein model systems, the NanoLuc enzyme and the WW structural motif. Both BayesDesign and the baseline ProteinMPNN algorithm increase the thermostability of NanoLuc and increase the conformational specificity of WW. The possible sources of error in the model are analyzed.

TAPIR: a T-cell receptor language model for predicting rare and novel targets

T-cell receptors (TCRs) are involved in most human diseases, but linking their sequences with their targets remains an unsolved grand challenge in the field. In this study, we present TAPIR (T-cell receptor and Peptide Interaction Recognizer), a T-cell receptor (TCR) language model that predicts TCR-target interactions, with a focus on novel and rare targets. TAPIR employs deep convolutional neural network (CNN) encoders to process TCR and target sequences across flexible representations (e.g., beta-chain only, unknown MHC allele, etc.) and learns patterns of interactivity via several training tasks. This flexibility allows TAPIR to train on more than 50k either paired (alpha and beta chain) or unpaired TCRs (just alpha or beta chain) from public and proprietary databases against 1933 unique targets. TAPIR demonstrates state-of-the-art performance when predicting TCR interactivity against common benchmark targets and is the first method to demonstrate strong performance when predicting TCR interactivity against novel targets, where no examples are provided in training. TAPIR is also capable of predicting TCR interaction against MHC alleles in the absence of target information. Leveraging these capabilities, we apply TAPIR to cancer patient TCR repertoires and identify and validate a novel and potent anti-cancer T-cell receptor against a shared cancer neoantigen target (PIK3CA H1047L). We further show how TAPIR, when extended with a generative neural network, is capable of directly designing T-cell receptor sequences that interact with a target of interest.

Design of stimulus-responsive two-state hinge proteins

In nature, proteins that switch between two conformations in response to environmental stimuli structurally transduce biochemical information in a manner analogous to how transistors control information flow in computing devices. Designing proteins with two distinct but fully structured conformations is a challenge for protein design as it requires sculpting an energy landscape with two distinct minima. Here we describe the design of "hinge" proteins that populate one designed state in the absence of ligand and a second designed state in the presence of ligand. X-ray crystallography, electron microscopy, double electron-electron resonance spectroscopy, and binding measurements demonstrate that despite the significant structural differences the two states are designed with atomic level accuracy and that the conformational and binding equilibria are closely coupled.

Are fibrinaloid microclots a cause of autoimmunity in Long Covid and other post-infection diseases?

It is now well established that the blood-clotting protein fibrinogen can polymerise into an anomalous form of fibrin that is amyloid in character; the resultant clots and microclots entrap many other molecules, stain with fluorogenic amyloid stains, are rather resistant to fibrinolysis, can block up microcapillaries, are implicated in a variety of diseases including Long COVID, and have been referred to as fibrinaloids. A necessary corollary of this anomalous polymerisation is the generation of novel epitopes in proteins that would normally be seen as 'self', and otherwise immunologically silent. The precise conformation of the resulting fibrinaloid clots (that, as with prions and classical amyloid proteins, can adopt multiple, stable conformations) must depend on the existing small molecules and metal ions that the fibrinogen may (and is some cases is known to) have bound before polymerisation. Any such novel epitopes, however, are likely to lead to the generation of autoantibodies. A convergent phenomenology, including distinct conformations and seeding of the anomalous form for initiation and propagation, is emerging to link knowledge in prions, prionoids, amyloids and now fibrinaloids. We here summarise the evidence for the above reasoning, which has substantial implications for our understanding of the genesis of autoimmunity (and the possible prevention thereof) based on the primary process of fibrinaloid formation.

Rapid and automated design of two-component protein nanomaterials using ProteinMPNN

The design of novel protein-protein interfaces using physics-based design methods such as Rosetta requires substantial computational resources and manual refinement by expert structural biologists. A new generation of deep learning methods promises to simplify protein-protein interface design and enable its application to a wide variety of problems by researchers from various scientific disciplines. Here we test the ability of a deep learning method for protein sequence design, ProteinMPNN, to design two-component tetrahedral protein nanomaterials and benchmark its performance against Rosetta. ProteinMPNN had a similar success rate to Rosetta, yielding 13 new experimentally confirmed assemblies, but required orders of magnitude less computation and no manual refinement. The interfaces designed by ProteinMPNN were substantially more polar than those designed by Rosetta, which facilitated in vitro assembly of the designed nanomaterials from independently purified components. Crystal structures of several of the assemblies confirmed the accuracy of the design method at high resolution. Our results showcase the potential of deep learning-based methods to unlock the widespread application of designed protein-protein interfaces and self-assembling protein nanomaterials in biotechnology.

De novo design of stable proteins that efficaciously inhibit oncogenic G proteins

Many protein therapeutics are competitive inhibitors that function by binding to endogenous proteins and preventing them from interacting with native partners. One effective strategy for engineering competitive inhibitors is to graft structural motifs from a native partner into a host protein. Here, we develop and experimentally test a computational protocol for embedding binding motifs in de novo designed proteins. The protocol uses an "inside-out" approach: Starting with a structural model of the binding motif docked against the target protein, the de novo protein is built by growing new structural elements off the termini of the binding motif. During backbone assembly, a score function favors backbones that introduce new tertiary contacts within the designed protein and do not introduce clashes with the target binding partner. Final sequences are designed and optimized using the molecular modeling program Rosetta. To test our protocol, we designed small helical proteins to inhibit the interaction between Gαq and its effector PLC-β isozymes. Several of the designed proteins remain folded above 90°C and bind to Gαq with equilibrium dissociation constants tighter than 80 nM. In cellular assays with oncogenic variants of Gαq , the designed proteins inhibit activation of PLC-β isozymes and Dbl-family RhoGEFs. Our results demonstrate that computational protein design, in combination with motif grafting, can be used to directly generate potent inhibitors without further optimization via high throughput screening or selection.

De novo design of protein structure and function with RFdiffusion

There has been considerable recent progress in designing new proteins using deep-learning methods1-9. Despite this progress, a general deep-learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher-order symmetric architectures, has yet to be described. Diffusion models10,11 have had considerable success in image and language generative modelling but limited success when applied to protein modelling, probably due to the complexity of protein backbone geometry and sequence-structure relationships. Here we show that by fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. The accuracy of RFdiffusion is confirmed by the cryogenic electron microscopy structure of a designed binder in complex with influenza haemagglutinin that is nearly identical to the design model. In a manner analogous to networks that produce images from user-specified inputs, RFdiffusion enables the design of diverse functional proteins from simple molecular specifications.

De Novo Design of Polyhedral Protein Assemblies: Before and After the AI Revolution

Self-assembling polyhedral protein biomaterials have gained attention as engineering targets owing to their naturally evolved sophisticated functions, ranging from protecting macromolecules from the environment to spatially controlling biochemical reactions. Precise computational design of de novo protein polyhedra is possible through two main types of approaches: methods from first principles, using physical and geometrical rules, and more recent data-driven methods based on artificial intelligence (AI), including deep learning (DL). Here, we retrospect first principle- and AI-based approaches for designing finite polyhedral protein assemblies, as well as advances in the structure prediction of such assemblies. We further highlight the possible applications of these materials and explore how the presented approaches can be combined to overcome current challenges and to advance the design of functional protein-based biomaterials.

Diagnosis, Treatment, and Management of Otitis Media with Artificial Intelligence

A common infectious disease, otitis media (OM) has a low rate of early diagnosis, which significantly increases the difficulty of treating the disease and the likelihood of serious complications developing including hearing loss, speech impairment, and even intracranial infection. Several areas of healthcare have shown great promise in the application of artificial intelligence (AI) systems, such as the accurate detection of diseases, the automated interpretation of images, and the prediction of patient outcomes. Several articles have reported some machine learning (ML) algorithms such as ResNet, InceptionV3 and Unet, were applied to the diagnosis of OM successfully. The use of these techniques in the OM is still in its infancy, but their potential is enormous. We present in this review important concepts related to ML and AI, describe how these technologies are currently being applied to diagnosing, treating, and managing OM, and discuss the challenges associated with developing AI-assisted OM technologies in the future.

Atomically Accurate Design of Metalloproteins with Predefined Coordination Geometries

We report a new computational protein design method for the construction of oligomeric protein assemblies around metal centers with predefined coordination geometries. We apply this method to design two homotrimeric assemblies, Tet4 and TP1, with tetrahedral and trigonal-pyramidal tris(histidine) metal coordination geometries, respectively, and demonstrate that both assemblies form the targeted metal centers with ≤0.2 Å accuracy. Although Tet4 and TP1 are constructed from the same parent protein building block, they are distinct in terms of their overall architectures, the environment surrounding the metal centers, and their metal-based reactivities, illustrating the versatility of our approach.

De novo designed ice-binding proteins from twist-constrained helices

Attaining molecular-level control over solidification processes is a crucial aspect of materials science. To control ice formation, organisms have evolved bewildering arrays of ice-binding proteins (IBPs), but these have poorly understood structure-activity relationships. We propose that reverse engineering using de novo computational protein design can shed light on structure-activity relationships of IBPs. We hypothesized that the model alpha-helical winter flounder antifreeze protein uses an unusual undertwisting of its alpha-helix to align its putative ice-binding threonine residues in exactly the same direction. We test this hypothesis by designing a series of straight three-helix bundles with an ice-binding helix projecting threonines and two supporting helices constraining the twist of the ice-binding helix. Our findings show that ice-recrystallization inhibition by the designed proteins increases with the degree of designed undertwisting, thus validating our hypothesis, and opening up avenues for the computational design of IBPs.

AlphaFold2 and Deep Learning for Elucidating Enzyme Conformational Flexibility and Its Application for Design

The recent success of AlphaFold2 (AF2) and other deep learning (DL) tools in accurately predicting the folded three-dimensional (3D) structure of proteins and enzymes has revolutionized the structural biology and protein design fields. The 3D structure indeed reveals key information on the arrangement of the catalytic machinery of enzymes and which structural elements gate the active site pocket. However, comprehending enzymatic activity requires a detailed knowledge of the chemical steps involved along the catalytic cycle and the exploration of the multiple thermally accessible conformations that enzymes adopt when in solution. In this Perspective, some of the recent studies showing the potential of AF2 in elucidating the conformational landscape of enzymes are provided. Selected examples of the key developments of AF2-based and DL methods for protein design are discussed, as well as a few enzyme design cases. These studies show the potential of AF2 and DL for allowing the routine computational design of efficient enzymes.

Small molecule allosteric modulation of the adenosine A1 receptor

G protein-coupled receptors (GPCRs) represent the target for approximately a third of FDA-approved small molecule drugs. The adenosine A₁ receptor (A₁R), one of four adenosine GPCR subtypes, has important (patho)physiological roles in humans. A₁R has well-established roles in the regulation of the cardiovascular and nervous systems, where it has been identified as a potential therapeutic target for a number of conditions, including cardiac ischemia-reperfusion injury, cognition, epilepsy, and neuropathic pain. A₁R small molecule drugs, typically orthosteric ligands, have undergone clinical trials. To date, none have progressed into the clinic, predominantly due to dose-limiting unwanted effects. The development of A₁R allosteric modulators that target a topographically distinct binding site represent a promising approach to overcome current limitations. Pharmacological parameters of allosteric ligands, including affinity, efficacy and cooperativity, can be optimized to regulate A₁R activity with high subtype, spatial and temporal selectivity. This review aims to offer insights into the A₁R as a potential therapeutic target and highlight recent advances in the structural understanding of A₁R allosteric modulation.

Alcohol acyltransferases for the biosynthesis of esters

Esters are widely used in food, energy, spices, chemical industry, etc., becoming an indispensable part of life. However, their production heavily relies on the fossil energy industry, which presents significant challenges associated with energy shortages and environmental pollution. Consequently, there is an urgent need to identify alternative green methods for ester production. One promising solution is biosynthesis, which offers sustainable and environmentally friendly processes. In ester biosynthesis, alcohol acyltransferases (AATs) catalyze the condensation of acyl-CoAs and alcohols to form esters, enabling the biosynthesis of nearly 100 different kinds of esters, such as ethyl acetate, hexyl acetate, ethyl crotonate, isoamyl acetate, and butyl butyrate. However, low catalytic efficiency and low selectivity of AATs represent the major bottlenecks for the biosynthesis of certain specific esters, which should be addressed with protein molecular engineering approaches before practical biotechnological applications. This review provides an overview of AAT enzymes, including their sequences, structures, active sites, catalytic mechanisms, and metabolic engineering applications. Furthermore, considering the critical role of AATs in determining the final ester products, the current research progresses of AAT modification using protein molecular engineering are also discussed. This review summarized the major challenges and prospects of AAT enzymes in ester biosynthesis.

An all-atom protein generative model

Proteins mediate their functions through chemical interactions; modeling these interactions, which are typically through sidechains, is an important need in protein design. However, constructing an all-atom generative model requires an appropriate scheme for managing the jointly continuous and discrete nature of proteins encoded in the structure and sequence. We describe an all-atom diffusion model of protein structure, Protpardelle, which instantiates a "superposition" over the possible sidechain states, and collapses it to conduct reverse diffusion for sample generation. When combined with sequence design methods, our model is able to co-design all-atom protein structure and sequence. Generated proteins are of good quality under the typical quality, diversity, and novelty metrics, and sidechains reproduce the chemical features and behavior of natural proteins. Finally, we explore the potential of our model conduct all-atom protein design and scaffold functional motifs in a backbone- and rotamer-free way.

Recent breakthroughs in computational structural biology harnessing the power of sequences and structures

Recent advances in computational approaches and their integration into structural biology enable tackling increasingly complex questions. Here, we discuss several key areas, highlighting breakthroughs and remaining challenges. Theoretical modeling has provided tools to accurately predict and design protein structures on a scale currently difficult to achieve using experimental approaches. Molecular Dynamics simulations have become faster and more precise, delivering actionable information inaccessible by current experimental methods. Virtual screening workflows allow a high-throughput approach to discover ligands that bind and modulate protein function, while Machine Learning methods enable the design of proteins with new functionalities. Integrative structural biology combines several of these approaches, pushing the frontiers of structural and functional characterization to ever larger systems, advancing towards a complete understanding of the living cell. These breakthroughs will accelerate and significantly impact diverse areas of science.

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.

De novo design of protein interactions with learned surface fingerprints

Physical interactions between proteins are essential for most biological processes governing life1. However, the molecular determinants of such interactions have been challenging to understand, even as genomic, proteomic and structural data increase. This knowledge gap has been a major obstacle for the comprehensive understanding of cellular protein-protein interaction networks and for the de novo design of protein binders that are crucial for synthetic biology and translational applications2-9. Here we use a geometric deep-learning framework operating on protein surfaces that generates fingerprints to describe geometric and chemical features that are critical to drive protein-protein interactions10. We hypothesized that these fingerprints capture the key aspects of molecular recognition that represent a new paradigm in the computational design of novel protein interactions. As a proof of principle, we computationally designed several de novo protein binders to engage four protein targets: SARS-CoV-2 spike, PD-1, PD-L1 and CTLA-4. Several designs were experimentally optimized, whereas others were generated purely in silico, reaching nanomolar affinity with structural and mutational characterization showing highly accurate predictions. Overall, our surface-centric approach captures the physical and chemical determinants of molecular recognition, enabling an approach for the de novo design of protein interactions and, more broadly, of artificial proteins with function.

Score-based generative modeling for de novo protein design

The generation of de novo protein structures with predefined functions and properties remains a challenging problem in protein design. Diffusion models, also known as score-based generative models (SGMs), have recently exhibited astounding empirical performance in image synthesis. Here we use image-based representations of protein structure to develop ProteinSGM, a score-based generative model that produces realistic de novo proteins. Through unconditional generation, we show that ProteinSGM can generate native-like protein structures, surpassing the performance of previously reported generative models. We experimentally validate some de novo designs and observe secondary structure compositions consistent with generated backbones. Finally, we apply conditional generation to de novo protein design by formulating it as an image inpainting problem, allowing precise and modular design of protein structure.

Fast, accurate antibody structure prediction from deep learning on massive set of natural antibodies

Antibodies have the capacity to bind a diverse set of antigens, and they have become critical therapeutics and diagnostic molecules. The binding of antibodies is facilitated by a set of six hypervariable loops that are diversified through genetic recombination and mutation. Even with recent advances, accurate structural prediction of these loops remains a challenge. Here, we present IgFold, a fast deep learning method for antibody structure prediction. IgFold consists of a pre-trained language model trained on 558 million natural antibody sequences followed by graph networks that directly predict backbone atom coordinates. IgFold predicts structures of similar or better quality than alternative methods (including AlphaFold) in significantly less time (under 25 s). Accurate structure prediction on this timescale makes possible avenues of investigation that were previously infeasible. As a demonstration of IgFold's capabilities, we predicted structures for 1.4 million paired antibody sequences, providing structural insights to 500-fold more antibodies than have experimentally determined structures.

Top-down design of protein architectures with reinforcement learning

As a result of evolutionary selection, the subunits of naturally occurring protein assemblies often fit together with substantial shape complementarity to generate architectures optimal for function in a manner not achievable by current design approaches. We describe a "top-down" reinforcement learning-based design approach that solves this problem using Monte Carlo tree search to sample protein conformers in the context of an overall architecture and specified functional constraints. Cryo-electron microscopy structures of the designed disk-shaped nanopores and ultracompact icosahedra are very close to the computational models. The icosohedra enable very-high-density display of immunogens and signaling molecules, which potentiates vaccine response and angiogenesis induction. Our approach enables the top-down design of complex protein nanomaterials with desired system properties and demonstrates the power of reinforcement learning in protein design.

Recent advances in predicting and modeling protein-protein interactions

Protein-protein interactions (PPIs) drive biological processes, and disruption of PPIs can cause disease. With recent breakthroughs in structure prediction and a deluge of genomic sequence data, computational methods to predict PPIs and model spatial structures of protein complexes are now approaching the accuracy of experimental approaches for permanent interactions and show promise for elucidating transient interactions. As we describe here, the key to this success is rich evolutionary information deciphered from thousands of homologous sequences that coevolve in interacting partners. This covariation signal, revealed by sophisticated statistical and machine learning (ML) algorithms, predicts physiological interactions. Accurate artificial intelligence (AI)-based modeling of protein structures promises to provide accurate 3D models of PPIs at a proteome-wide scale.

In Silico Approaches for Prediction of Anti-CRISPR Proteins

Numerous viruses infecting bacteria and archaea encode CRISPR-Cas system inhibitors, known as anti-CRISPR proteins (Acr). The Acrs typically are highly specific for particular CRISPR variants, resulting in remarkable sequence and structural diversity and complicating accurate prediction and identification of Acrs. In addition to their intrinsic interest for understanding the coevolution of defense and counter-defense systems in prokaryotes, Acrs could be natural, potent on-off switches for CRISPR-based biotechnological tools, so their discovery, characterization and application are of major importance. Here we discuss the computational approaches for Acr prediction. Due to the enormous diversity and likely multiple origins of the Acrs, sequence similarity searches are of limited use. However, multiple features of protein and gene organization have been successfully harnessed to this end including small protein size and distinct amino acid compositions of the Acrs, association of acr genes in virus genomes with genes encoding helix-turn-helix proteins that regulate Acr expression (Acr-associated proteins, Aca), and presence of self-targeting CRISPR spacers in bacterial and archaeal genomes containing Acr-encoding proviruses. Productive approaches for Acr prediction also involve genome comparison of closely related viruses, of which one is resistant and the other one is sensitive to a particular CRISPR variant, and "guilt by association" whereby genes adjacent to a homolog of a known Aca are identified as candidate Acrs. The distinctive features of Acrs are employed for Acr prediction both by developing dedicated search algorithms and through machine learning. New approaches will be needed to identify novel types of Acrs that are likely to exist.

De novo design of stable proteins that efficaciously inhibit oncogenic G proteins

Many protein therapeutics are competitive inhibitors that function by binding to endogenous proteins and preventing them from interacting with native partners. One effective strategy for engineering competitive inhibitors is to graft structural motifs from a native partner into a host protein. Here, we develop and experimentally test a computational protocol for embedding binding motifs in de novo designed proteins. The protocol uses an "inside-out" approach: Starting with a structural model of the binding motif docked against the target protein, the de novo protein is built by growing new structural elements off the termini of the binding motif. During backbone assembly, a score function favors backbones that introduce new tertiary contacts within the designed protein and do not introduce clashes with the target binding partner. Final sequences are designed and optimized using the molecular modeling program Rosetta. To test our protocol, we designed small helical proteins to inhibit the interaction between Gα q and its effector PLC-β isozymes. Several of the designed proteins remain folded above 90°C and bind to Gα q with equilibrium dissociation constants tighter than 80 nM. In cellular assays with oncogenic variants of Gα q , the designed proteins inhibit activation of PLC-β isozymes and Dbl-family RhoGEFs. Our results demonstrate that computational protein design, in combination with motif grafting, can be used to directly generate potent inhibitors without further optimization via high throughput screening or selection.

statement for broader audience

Engineered proteins that bind to specific target proteins are useful as research reagents, diagnostics, and therapeutics. We used computational protein design to engineer de novo proteins that bind and competitively inhibit the G protein, Gα q , which is an oncogene for uveal melanomas. This computational method is a general approach that should be useful for designing competitive inhibitors against other proteins of interest.

Super High-Throughput Screening of Enzyme Variants by Spectral Graph Convolutional Neural Networks

Finding new enzyme variants with the desired substrate scope requires screening through a large number of potential variants. In a typical in silico enzyme engineering workflow, it is possible to scan a few thousands of variants, and gather several candidates for further screening or experimental verification. In this work, we show that a Graph Convolutional Neural Network (GCN) can be trained to predict the binding energy of combinatorial libraries of enzyme complexes using only sequence information. The GCN model uses a stack of message-passing and graph pooling layers to extract information from the protein input graph and yield a prediction. The GCN model is agnostic to the identity of the ligand, which is kept constant within the mutant libraries. Using a miniscule subset of the total combinatorial space (20⁴-20⁸ mutants) as training data, the proposed GCN model achieves a high accuracy in predicting the binding energy of unseen variants. The network's accuracy was further improved by injecting feature embeddings obtained from a language module pretrained on 10 million protein sequences. Since no structural information is needed to evaluate new variants, the deep learning algorithm is capable of scoring an enzyme variant in under 1 ms, allowing the search of billions of candidates on a single GPU.

PSL Chemical Biology Symposia Third Edition: A Branch of Science in its Explosive Phase

This symposium is the third PSL (Paris Sciences & Lettres) Chemical Biology meeting (2016, 2019, 2023) held at Institut Curie. This initiative originally started at Institut de Chimie des Substances Naturelles (ICSN) in Gif-sur-Yvette (2013, 2014), under the directorship of Professor Max Malacria, with a strong focus on chemistry. It was then continued at the Institut Curie (2015) covering a larger scope, before becoming the official PSL Chemical Biology meeting. This latest edition was postponed twice for the reasons that we know. This has given us the opportunity to invite additional speakers of great standing. This year, Institut Curie hosted around 300 participants, including 220 on site and over 80 online. The pandemic has had, at least, the virtue of promoting online meetings, which we came to realize is not perfect but has its own merits. In particular, it enables those with restricted time and resources to take part in events and meetings, which can now accommodate unlimited participants. We apologize to all those who could not attend in person this time due to space limitation at Institut Curie.

AsiteDesign: a Semirational Algorithm for an Automated Enzyme Design

With advances in protein structure predictions, the number of available high-quality structures has increased dramatically. In light of these advances, structure-based enzyme engineering is expected to become increasingly important for optimizing biocatalysts for industrial processes. Here, we present AsiteDesign, a Monte Carlo-based protocol for structure-based engineering of active sites. AsiteDesign provides a framework for introducing new catalytic residues in a given binding pocket to either create a new catalytic activity or alter the existing one. AsiteDesign is implemented using pyRosetta and incorporates enhanced sampling techniques to efficiently explore the search space. The protocol was tested by designing an alternative catalytic triad in the active site of Pseudomonas fluorescens esterase (PFE). The designed variant was experimentally verified to be active, demonstrating that AsiteDesign can find alternative catalytic triads. Additionally, the AsiteDesign protocol was employed to enhance the hydrolysis of a bulky chiral substrate (1-phenyl-2-pentyl acetate) by PFE. The experimental verification of the designed variants demonstrated that F158L/F198A and F125A/F158L mutations increased the hydrolysis of 1-phenyl-2-pentyl acetate from 8.9 to 66.7 and 23.4%, respectively, and reversed the enantioselectivity of the enzyme from (R) to (S)-enantiopreference, with 32 and 55% enantiomeric excess (ee), respectively.

Machine learning for evolutionary-based and physics-inspired protein design: Current and future synergies

Computational protein design facilitates the discovery of novel proteins with prescribed structure and functionality. Exciting designs were recently reported using novel data-driven methodologies that can be roughly divided into two categories: evolutionary-based and physics-inspired approaches. The former infer characteristic sequence features shared by sets of evolutionary-related proteins, such as conserved or coevolving positions, and recombine them to generate candidates with similar structure and function. The latter approaches estimate key biochemical properties, such as structure free energy, conformational entropy, or binding affinities using machine learning surrogates, and optimize them to yield improved designs. Here, we review recent progress along both tracks, discuss their strengths and weaknesses, and highlight opportunities for synergistic approaches.

Evolutionary-scale prediction of atomic-level protein structure with a language model

Recent advances in machine learning have leveraged evolutionary information in multiple sequence alignments to predict protein structure. We demonstrate direct inference of full atomic-level protein structure from primary sequence using a large language model. As language models of protein sequences are scaled up to 15 billion parameters, an atomic-resolution picture of protein structure emerges in the learned representations. This results in an order-of-magnitude acceleration of high-resolution structure prediction, which enables large-scale structural characterization of metagenomic proteins. We apply this capability to construct the ESM Metagenomic Atlas by predicting structures for >617 million metagenomic protein sequences, including >225 million that are predicted with high confidence, which gives a view into the vast breadth and diversity of natural proteins.

Determinants for an Efficient Enzymatic Catalysis in Poly(Ethylene Terephthalate) Degradation

The enzymatic degradation of the recalcitrant poly(ethylene terephthalate) (PET) has been an important biotechnological goal. The present review focuses on the state of the art in enzymatic degradation of PET, and the challenges ahead. This review covers (i) enzymes acting on PET, (ii) protein improvements through selection or engineering, (iii) strategies to improve biocatalyst–polymer interaction and monomer yields. Finally, this review discusses critical points on PET degradation, and their related experimental aspects, that include the control of physicochemical parameters. The search for, and engineering of, PET hydrolases, have been widely studied to achieve this, and several examples are discussed here. Many enzymes, from various microbial sources, have been studied and engineered, but recently true PET hydrolases (PETases), active at moderate temperatures, were reported. For a circular economy process, terephtalic acid (TPA) production is critical. Some thermophilic cutinases and engineered PETases have been reported to release terephthalic acid in significant amounts. Some bottlenecks in enzyme performance are discussed, including enzyme activity, thermal stability, substrate accessibility, PET microstructures, high crystallinity, molecular mass, mass transfer, and efficient conversion into reusable fragments.

AlphaFold2 and its applications in the fields of biology and medicine

AlphaFold2 (AF2) is an artificial intelligence (AI) system developed by DeepMind that can predict three-dimensional (3D) structures of proteins from amino acid sequences with atomic-level accuracy. Protein structure prediction is one of the most challenging problems in computational biology and chemistry, and has puzzled scientists for 50 years. The advent of AF2 presents an unprecedented progress in protein structure prediction and has attracted much attention. Subsequent release of structures of more than 200 million proteins predicted by AF2 further aroused great enthusiasm in the science community, especially in the fields of biology and medicine. AF2 is thought to have a significant impact on structural biology and research areas that need protein structure information, such as drug discovery, protein design, prediction of protein function, et al. Though the time is not long since AF2 was developed, there are already quite a few application studies of AF2 in the fields of biology and medicine, with many of them having preliminarily proved the potential of AF2. To better understand AF2 and promote its applications, we will in this article summarize the principle and system architecture of AF2 as well as the recipe of its success, and particularly focus on reviewing its applications in the fields of biology and medicine. Limitations of current AF2 prediction will also be discussed.

Exploring and Learning the Universe of Protein Allostery Using Artificial Intelligence Augmented Biophysical and Computational Approaches

Allosteric mechanisms are commonly employed regulatory tools used by proteins to orchestrate complex biochemical processes and control communications in cells. The quantitative understanding and characterization of allosteric molecular events are among major challenges in modern biology and require integration of innovative computational experimental approaches to obtain atomistic-level knowledge of the allosteric states, interactions, and dynamic conformational landscapes. The growing body of computational and experimental studies empowered by emerging artificial intelligence (AI) technologies has opened up new paradigms for exploring and learning the universe of protein allostery from first principles. In this review we analyze recent developments in high-throughput deep mutational scanning of allosteric protein functions; applications and latest adaptations of Alpha-fold structural prediction methods for studies of protein dynamics and allostery; new frontiers in integrating machine learning and enhanced sampling techniques for characterization of allostery; and recent advances in structural biology approaches for studies of allosteric systems. We also highlight recent computational and experimental studies of the SARS-CoV-2 spike (S) proteins revealing an important and often hidden role of allosteric regulation driving functional conformational changes, binding interactions with the host receptor, and mutational escape mechanisms of S proteins which are critical for viral infection. We conclude with a summary and outlook of future directions suggesting that AI-augmented biophysical and computer simulation approaches are beginning to transform studies of protein allostery toward systematic characterization of allosteric landscapes, hidden allosteric states, and mechanisms which may bring about a new revolution in molecular biology and drug discovery.

Protein Design Using Physics Informed Neural Networks

The inverse protein folding problem, also known as protein sequence design, seeks to predict an amino acid sequence that folds into a specific structure and performs a specific function. Recent advancements in machine learning techniques have been successful in generating functional sequences, outperforming previous energy function-based methods. However, these machine learning methods are limited in their interoperability and robustness, especially when designing proteins that must function under non-ambient conditions, such as high temperature, extreme pH, or in various ionic solvents. To address this issue, we propose a new Physics-Informed Neural Networks (PINNs)-based protein sequence design approach. Our approach combines all-atom molecular dynamics simulations, a PINNs MD surrogate model, and a relaxation of binary programming to solve the protein design task while optimizing both energy and the structural stability of proteins. We demonstrate the effectiveness of our design framework in designing proteins that can function under non-ambient conditions.

Cyclic peptide structure prediction and design using AlphaFold

Deep learning networks offer considerable opportunities for accurate structure prediction and design of biomolecules. While cyclic peptides have gained significant traction as a therapeutic modality, developing deep learning methods for designing such peptides has been slow, mostly due to the small number of available structures for molecules in this size range. Here, we report approaches to modify the AlphaFold network for accurate structure prediction and design of cyclic peptides. Our results show this approach can accurately predict the structures of native cyclic peptides from a single sequence, with 36 out of 49 cases predicted with high confidence (pLDDT > 0.85) matching the native structure with root mean squared deviation (RMSD) less than 1.5 Å. Further extending our approach, we describe computational methods for designing sequences of peptide backbones generated by other backbone sampling methods and for de novo design of new macrocyclic peptides. We extensively sampled the structural diversity of cyclic peptides between 7-13 amino acids, and identified around 10,000 unique design candidates predicted to fold into the designed structures with high confidence. X-ray crystal structures for seven sequences with diverse sizes and structures designed by our approach match very closely with the design models (root mean squared deviation < 1.0 Å), highlighting the atomic level accuracy in our approach. The computational methods and scaffolds developed here provide the basis for custom-designing peptides for targeted therapeutic applications.

Possibilities of Using De Novo Design for Generating Diverse Functional Food Enzymes

Food enzymes have an important role in the improvement of certain food characteristics, such as texture improvement, elimination of toxins and allergens, production of carbohydrates, enhancing flavor/appearance characteristics. Recently, along with the development of artificial meats, food enzymes have been employed to achieve more diverse functions, especially in converting non-edible biomass to delicious foods. Reported food enzyme modifications for specific applications have highlighted the significance of enzyme engineering. However, using direct evolution or rational design showed inherent limitations due to the mutation rates, which made it difficult to satisfy the stability or specific activity needs for certain applications. Generating functional enzymes using de novo design, which highly assembles naturally existing enzymes, provides potential solutions for screening desired enzymes. Here, we describe the functions and applications of food enzymes to introduce the need for food enzymes engineering. To illustrate the possibilities of using de novo design for generating diverse functional proteins, we reviewed protein modelling and de novo design methods and their implementations. The future directions for adding structural data for de novo design model training, acquiring diversified training data, and investigating the relationship between enzyme-substrate binding and activity were highlighted as challenges to overcome for the de novo design of food enzymes.

Inside-Out Design of Zinc-Binding Proteins with Non-Native Backbones

The de novo design of functional proteins requires specification of tertiary structure and incorporation of molecular binding sites. Here, we develop an inside-out design strategy in the molecular modeling program Rosetta that begins with amino acid side chains from one or two α-helices making well-defined contacts with a ligand. A full-sized protein is then built around the ligand by adding additional helices that promote the formation of a protein core and allow additional contacts with the ligand. The protocol was tested by designing 12 zinc-binding proteins, each with 4-5 helices. Four of the designs were folded and bound to zinc with equilibrium dissociation constants varying between 95 nM and 1.1 μM. The design with the tightest affinity for zinc, N12, adopts a unique conformation in the folded state as assessed with nuclear magnetic resonance (NMR) and the design model closely matches (backbone root-mean-square deviation (RMSD) < 1 Å) an AlphaFold model of the sequence. Retrospective analysis with AlphaFold suggests that the sequences of many of the failed designs did not encode the desired tertiary packing.

De Novo Design and Synthesis of Polypeptide Immunomodulators for Resetting Macrophage Polarization

Modulating the extracellular matrix microenvironment is critical for achieving the desired macrophage phenotype in immune investigations or tumor therapy. Combining de novo protein design and biosynthesis techniques, herein, we designed a biomimetic polypeptide self-assembled nano-immunomodulator to trigger the activation of a specific macrophage phenotype. It was intended to be made up of (​GGS​GGP​GGG​PAS​AAA​NSA​SRA​TSN​SP)n, the RGD motif from collagen, and the IKVAV motif from laminin. The combination of these domains allows the biomimetic polypeptide to assemble into extracellular matrix-like nanofibrils, creating an extracellular matrix-like milieu for macrophages. Furthermore, changing the concentration further provides a facile route to fine-tune macrophage polarization, which enhances antitumor immune responses by precisely resetting tumor-associated macrophage immune responses into an M1-like phenotype, which is generally considered to be tumor-killing macrophages, primarily antitumor, and immune-promoting. Unlike metal or synthetic polymer-based nanoparticles, this polypeptide-based nanomaterial exhibits excellent biocompatibility, high efficacy, and precise tunability in immunomodulatory effectiveness. These encouraging findings motivate us to continue our research into cancer immunotherapy applications in the future.

Ranking Peptide Binders by Affinity with AlphaFold

AlphaFold has revolutionized structural biology by predicting highly accurate structures of proteins and their complexes with peptides and other proteins. However, for protein-peptide systems, we are also interested in identifying the highest affinity binder among a set of candidate peptides. We present a novel competitive binding assay using AlphaFold to predict structures of the receptor in the presence of two peptides. For systems in which the individual structures of the peptides are well predicted, the assay captures the higher affinity binder in the bound state, and the other peptide in the unbound form with statistical significance. We test the application on six protein receptors for which we have experimental binding affinities to several peptides. We find that the assay is best suited for identifying medium to strong peptide binders that adopt stable secondary structures upon binding.

Big data mining, rational modification, and ancestral sequence reconstruction inferred multiple xylose isomerases for biorefinery

The isomerization of xylose to xylulose is considered the most promising approach to initiate xylose bioconversion. Here, phylogeny-guided big data mining, rational modification, and ancestral sequence reconstruction strategies were implemented to explore new active xylose isomerases (XIs) for Saccharomyces cerevisiae. Significantly, 13 new active XIs for S. cerevisiae were mined or artificially created. Moreover, the importance of the amino-terminal fragment for maintaining basic XI activity was demonstrated. With the mined XIs, four efficient xylose-utilizing S. cerevisiae were constructed and evolved, among which the strain S. cerevisiae CRD5HS contributed to ethanol titers as high as 85.95 and 94.76 g/liter from pretreated corn stover and corn cob, respectively, without detoxifying or washing pretreated biomass. Potential genetic targets obtained from adaptive laboratory evolution were further analyzed by sequencing the high-performance strains. The combined XI mining methods described here provide practical references for mining other scarce and valuable enzymes.

Construction of Fusion Protein with Carbohydrate-Binding Module and Leaf-Branch Compost Cutinase to Enhance the Degradation Efficiency of Polyethylene Terephthalate

Poly(ethylene terephthalate) (PET) is a manufactured plastic broadly available, whereas improper disposal of PET waste has become a serious burden on the environment. Leaf-branch compost cutinase (LCC) is one of the most powerful and promising PET hydrolases, and its mutant LCCICCG shows high catalytic activity and excellent thermal stability. However, low binding affinity with PET has been found to dramatically limit its further industrial application. Herein, TrCBM and CfCBM were rationally selected from the CAZy database to construct fusion proteins with LCCICCG, and mechanistic studies revealed that these two domains could bind with PET favorably via polar amino acids. The optimal temperatures of LCCICCG-TrCBM and CfCBM-LCCICCG were measured to be 70 and 80 °C, respectively. Moreover, these two fusion proteins exhibited favorable thermal stability, maintaining 53.1% and 48.8% of initial activity after the incubation at 90 °C for 300 min. Compared with LCCICCG, the binding affinity of LCCICCG-TrCBM and CfCBM-LCCICCG for PET has been improved by 1.4- and 1.3-fold, respectively, and meanwhile their degradation efficiency on PET films was enhanced by 3.7% and 24.2%. Overall, this study demonstrated that the strategy of constructing fusion proteins is practical and prospective to facilitate the enzymatic PET degradation ability.

De novo design of luciferases using deep learning

De novo enzyme design has sought to introduce active sites and substrate-binding pockets that are predicted to catalyse a reaction of interest into geometrically compatible native scaffolds1,2, but has been limited by a lack of suitable protein structures and the complexity of native protein sequence-structure relationships. Here we describe a deep-learning-based 'family-wide hallucination' approach that generates large numbers of idealized protein structures containing diverse pocket shapes and designed sequences that encode them. We use these scaffolds to design artificial luciferases that selectively catalyse the oxidative chemiluminescence of the synthetic luciferin substrates diphenylterazine3 and 2-deoxycoelenterazine. The designed active sites position an arginine guanidinium group adjacent to an anion that develops during the reaction in a binding pocket with high shape complementarity. For both luciferin substrates, we obtain designed luciferases with high selectivity; the most active of these is a small (13.9 kDa) and thermostable (with a melting temperature higher than 95 °C) enzyme that has a catalytic efficiency on diphenylterazine (kcat/Km = 106 M-1 s-1) comparable to that of native luciferases, but a much higher substrate specificity. The creation of highly active and specific biocatalysts from scratch with broad applications in biomedicine is a key milestone for computational enzyme design, and our approach should enable generation of a wide range of luciferases and other enzymes.

De novo protein fold design through sequence-independent fragment assembly simulations

De novo protein design generally consists of two steps, including structure and sequence design. Many protein design studies have focused on sequence design with scaffolds adapted from native structures in the PDB, which renders novel areas of protein structure and function space unexplored. We developed FoldDesign to create novel protein folds from specific secondary structure (SS) assignments through sequence-independent replica-exchange Monte Carlo (REMC) simulations. The method was tested on 354 non-redundant topologies, where FoldDesign consistently created stable structural folds, while recapitulating on average 87.7% of the SS elements. Meanwhile, the FoldDesign scaffolds had well-formed structures with buried residues and solvent-exposed areas closely matching their native counterparts. Despite the high fidelity to the input SS restraints and local structural characteristics of native proteins, a large portion of the designed scaffolds possessed global folds completely different from natural proteins in the PDB, highlighting the ability of FoldDesign to explore novel areas of protein fold space. Detailed data analyses revealed that the major contributions to the successful structure design lay in the optimal energy force field, which contains a balanced set of SS packing terms, and REMC simulations, which were coupled with multiple auxiliary movements to efficiently search the conformational space. Additionally, the ability to recognize and assemble uncommon super-SS geometries, rather than the unique arrangement of common SS motifs, was the key to generating novel folds. These results demonstrate a strong potential to explore both structural and functional spaces through computational design simulations that natural proteins have not reached through evolution.

Rational design of enzyme activity and enantioselectivity

The strategy of rational design to engineer enzymes is to predict the potential mutants based on the understanding of the relationships between protein structure and function, and subsequently introduce the mutations using the site-directed mutagenesis. Rational design methods are universal, relatively fast and have the potential to be developed into algorithms that can quantitatively predict the performance of the designed sequences. Compared to the protein stability, it was more challenging to design an enzyme with improved activity or selectivity, due to the complexity of enzyme molecular structure and inadequate understanding of the relationships between enzyme structures and functions. However, with the development of computational force, advanced algorithm and a deeper understanding of enzyme catalytic mechanisms, rational design could significantly simplify the process of engineering enzyme functions and the number of studies applying rational design strategy has been increasing. Here, we reviewed the recent advances of applying the rational design strategy to engineer enzyme functions including activity and enantioselectivity. Five strategies including multiple sequence alignment, strategy based on steric hindrance, strategy based on remodeling interaction network, strategy based on dynamics modification and computational protein design are discussed and the successful cases using these strategies are introduced.