Deep learning to design nuclear-targeting abiotic miniproteins

Protein nanoparticles as drug delivery systems for cancer theranostics

Protein-based nanoparticles have garnered significant attention in theranostic applications due to their superior biocompatibility, exceptional biodegradability and ease of functionality. Compared to other nanocarriers, protein-based nanoparticles offer additional advantages, including biofunctionality and precise molecular recognition abilities, which make them highly effective in navigating complex biological environments. Moreover, proteins can serve as powerful tools with self-assembling structures and reagents that enhance cell penetration. And their derivation from abundant renewable sources and ability to degrade into harmless amino acids further enhance their suitability for biomedical applications. However, protein-based nanoparticles have so far not realized their full potential. In this review, we summarize recent advances in the use of protein nanoparticles in tumor diagnosis and treatment and outline typical methods for preparing protein nanoparticles. The review of protein nanoparticles may provide useful new insights into the development of biomaterial fabrication.

Design of Cytotoxic T Cell Epitopes by Machine Learning of Human Degrons

Antigen processing is critical for therapeutic vaccines to generate epitopes for priming cytotoxic T cell responses against cancer and pathogens, but insufficient processing often limits the quantity of epitopes released. We address this challenge using machine learning to ascribe a proteasomal degradation score to epitope sequences. Epitopes with varying scores were translocated into cells using nontoxic anthrax proteins. Epitopes with a low score show pronounced immunogenicity due to antigen processing, but epitopes with a high score show limited immunogenicity. This work sheds light on the sequence-activity relationships between proteasomal degradation and epitope immunogenicity. We anticipate that future efforts to incorporate proteasomal degradation signals into vaccine designs will lead to enhanced cytotoxic T cell priming by these vaccines in clinical settings.

Deep learning for advancing peptide drug development: Tools and methods in structure prediction and design

Peptides can bind challenging disease targets with high affinity and specificity, offering enormous opportunities for addressing unmet medical needs. However, peptides' unique features, including smaller size, increased structural flexibility, and limited data availability, pose additional challenges to the design process compared to proteins. This review explores the dynamic field of peptide therapeutics, leveraging deep learning to enhance structure prediction and design. Our exploration encompasses various facets of peptide research, ranging from dataset curation handling to model development. As deep learning technologies become more refined, we channel our efforts into peptide structure prediction and design, aligning with the fundamental principles of structure-activity relationships in drug development. To guide researchers in harnessing the potential of deep learning to advance peptide drug development, our insights comprehensively explore current challenges and future directions of peptide therapeutics.

Protein-based nanoparticles for therapeutic nucleic acid delivery

To realize the full potential of emerging nucleic acid therapies, there is a need for effective delivery agents to transport cargo to cells of interest. Protein materials exhibit several unique properties, including biodegradability, biocompatibility, ease of functionalization via recombinant and chemical modifications, among other features, which establish a promising basis for therapeutic nucleic acid delivery systems. In this review, we highlight progress made in the use of non-viral protein-based nanoparticles for nucleic acid delivery in vitro and in vivo, while elaborating on key physicochemical properties that have enabled the use of these materials for nanoparticle formulation and drug delivery. To conclude, we comment on the prospects and unresolved challenges associated with the translation of protein-based nucleic acid delivery systems for therapeutic applications.

Review on Uric Acid Recognition by MOFs with a Future in Machine Learning

Uric acid (UA) is produced from purine metabolism and serves as a prevalent biomarker for multiple diseases including cancer. Hyperuricemia or hypouricemia can cause multiple dysfunctions throughout the biological processes. Consequently, there is a pressing need for monitoring UA concentration in body fluid. While clinical methods are known, the availability of a point-of-care testing (PoCT) kit remains conspicuously absent. In the case of electrochemical recognition of UA, the oxidation potential of ascorbic acid closely aligns with that of UA and thus it hinders the detection process, which eventually may result in false positive signals. Several chemosensors are known in the field of supramolecular chemistry, and metal-organic frameworks (MOFs) are one of the best-performing contenders due to their robustness, stability, and versatile structures. In this review, we tried to unbox the up-to-date development of UA sensing by MOFs. We delve into the state of UA recognition by MOFs, exploring both electrochemical and fluorometric pathways and drawing comparisons with structurally similar probes like covalent organic frameworks (COFs) to understand/establish the advantages of MOFs specifically in UA sensing. In the absence of a PoCT kit, we have provided the conceptual outlook for designing a PoCT device termed a "Urimeter" via electrochemical operation. For the first time, we have proposed different methods of how UA sensing can be tied up with artificial intelligence and machine learning (AI-ML).

Design of Cytotoxic T Cell Epitopes by Machine Learning of Human Degrons

Antigen processing is critical for producing epitope peptides that are presented by HLA molecules for T cell recognition. Therapeutic vaccines aim to harness these epitopes for priming cytotoxic T cell responses against cancer and pathogens, but insufficient processing often reduces vaccine efficacy through limiting the quantity of epitopes released. Here, we set out to improve antigen processing by harnessing protein degradation signals called degrons from the ubiquitin-proteasome system. We used machine learning to generate a computational model that ascribes a proteasomal degradation score between 0 and 100. Epitope peptides with varying degron activities were synthesized and translocated into cells using nontoxic anthrax proteins: protective antigen (PA) and the N-terminus of lethal factor (LFN). Immunogenicity studies revealed epitope sequences with a low score (<25) show pronounced T-cell activation but epitope sequences with a higher score (>75) provide limited activation. This work sheds light on the sequence-activity relationships between proteasomal degradation and epitope immunogenicity, through conserving the epitope region but varying the flanking sequence. We anticipate that future efforts to incorporate proteasomal degradation signals into vaccine designs will lead to enhanced cytotoxic T cell priming by vaccine therapeutics in clinical settings.

In vitro continuous protein evolution empowered by machine learning and automation

Directed evolution has become one of the most successful and powerful tools for protein engineering. However, the efforts required for designing, constructing, and screening a large library of variants can be laborious, time-consuming, and costly. With the recent advent of machine learning (ML) in the directed evolution of proteins, researchers can now evaluate variants in silico and guide a more efficient directed evolution campaign. Furthermore, recent advancements in laboratory automation have enabled the rapid execution of long, complex experiments for high-throughput data acquisition in both industrial and academic settings, thus providing the means to collect a large quantity of data required to develop ML models for protein engineering. In this perspective, we propose a closed-loop in vitro continuous protein evolution framework that leverages the best of both worlds, ML and automation, and provide a brief overview of the recent developments in the field.

Accelerating therapeutic protein design with computational approaches toward the clinical stage

Therapeutic protein, represented by antibodies, is of increasing interest in human medicine. However, clinical translation of therapeutic protein is still largely hindered by different aspects of developability, including affinity and selectivity, stability and aggregation prevention, solubility and viscosity reduction, and deimmunization. Conventional optimization of the developability with widely used methods, like display technologies and library screening approaches, is a time and cost-intensive endeavor, and the efficiency in finding suitable solutions is still not enough to meet clinical needs. In recent years, the accelerated advancement of computational methodologies has ushered in a transformative era in the field of therapeutic protein design. Owing to their remarkable capabilities in feature extraction and modeling, the integration of cutting-edge computational strategies with conventional techniques presents a promising avenue to accelerate the progression of therapeutic protein design and optimization toward clinical implementation. Here, we compared the differences between therapeutic protein and small molecules in developability and provided an overview of the computational approaches applicable to the design or optimization of therapeutic protein in several developability issues.

A Systematic Review of Deep Learning Methodologies Used in the Drug Discovery Process with Emphasis on In Vivo Validation

The discovery and development of new drugs are extremely long and costly processes. Recent progress in artificial intelligence has made a positive impact on the drug development pipeline. Numerous challenges have been addressed with the growing exploitation of drug-related data and the advancement of deep learning technology. Several model frameworks have been proposed to enhance the performance of deep learning algorithms in molecular design. However, only a few have had an immediate impact on drug development since computational results may not be confirmed experimentally. This systematic review aims to summarize the different deep learning architectures used in the drug discovery process and are validated with further in vivo experiments. For each presented study, the proposed molecule or peptide that has been generated or identified by the deep learning model has been biologically evaluated in animal models. These state-of-the-art studies highlight that even if artificial intelligence in drug discovery is still in its infancy, it has great potential to accelerate the drug discovery cycle, reduce the required costs, and contribute to the integration of the 3R (Replacement, Reduction, Refinement) principles. Out of all the reviewed scientific articles, seven algorithms were identified: recurrent neural networks, specifically, long short-term memory (LSTM-RNNs), Autoencoders (AEs) and their Wasserstein Autoencoders (WAEs) and Variational Autoencoders (VAEs) variants; Convolutional Neural Networks (CNNs); Direct Message Passing Neural Networks (D-MPNNs); and Multitask Deep Neural Networks (MTDNNs). LSTM-RNNs were the most used architectures with molecules or peptide sequences as inputs.

In-Cell Penetration Selection-Mass Spectrometry Produces Noncanonical Peptides for Antisense Delivery

Peptide-mediated delivery of macromolecules in cells has significant potential therapeutic benefits, but no therapy employing cell-penetrating peptides (CPPs) has reached the market after 30 years of investigation due to challenges in the discovery of new, more efficient sequences. Here, we demonstrate a method for in-cell penetration selection-mass spectrometry (in-cell PS-MS) to discover peptides from a synthetic library capable of delivering macromolecule cargo to the cytosol. This method was inspired by recent in vivo selection approaches for cell-surface screening, with an added spatial dimension resulting from subcellular fractionation. A representative peptide discovered in the cytosolic extract, Cyto1a, is nearly 100-fold more active toward antisense phosphorodiamidate morpholino oligomer (PMO) delivery compared to a sequence identified from a whole cell extract, which includes endosomes. Cyto1a is composed of d-residues and two non-α-amino acids, is more stable than its all-l isoform, and is less toxic than known CPPs with comparable activity. Pulse-chase and microscopy experiments revealed that while the PMO-Cyto1a conjugate is likely taken up by endosomes, it can escape to localize to the nucleus without nonspecifically releasing other endosomal components. In-cell PS-MS introduces a means to empirically discover unnatural synthetic peptides for subcellular delivery of therapeutically relevant cargo.

In Silico Screening and Optimization of Cell-Penetrating Peptides Using Deep Learning Methods

Cell-penetrating peptides (CPPs) have great potential to deliver bioactive agents into cells. Although there have been many recent advances in CPP-related research, it is still important to develop more efficient CPPs. The development of CPPs by in silico methods is a very useful addition to experimental methods, but in many cases it can lead to a large number of false-positive results. In this study, we developed a deep-learning-based CPP prediction method, AiCPP, to develop novel CPPs. AiCPP uses a large number of peptide sequences derived from human-reference proteins as a negative set to reduce false-positive predictions and adopts a method to learn small-length peptide sequence motifs that may have CPP tendencies. Using AiCPP, we found that short peptide sequences derived from amyloid precursor proteins are efficient new CPPs, and experimentally confirmed that these CPP sequences can be further optimized.

Roles of Skeletal Muscle in Development: A Bioinformatics and Systems Biology Overview

The ability to assess various cellular events consequent to perturbations, such as genetic mutations, disease states and therapies, has been recently revolutionized by technological advances in multiple "omics" fields. The resulting deluge of information has enabled and necessitated the development of tools required to both process and interpret the data. While of tremendous value to basic researchers, the amount and complexity of the data has made it extremely difficult to manually draw inference and identify factors key to the study objectives. The challenges of data reduction and interpretation are being met by the development of increasingly complex tools that integrate disparate knowledge bases and synthesize coherent models based on current biological understanding. This chapter presents an example of how genomics data can be integrated with biological network analyses to gain further insight into the developmental consequences of genetic perturbations. State of the art methods for conducting similar studies are discussed along with modern methods used to analyze and interpret the data.

Machine Learning Guides Peptide Nucleic Acid Flow Synthesis and Sequence Design

Peptide nucleic acids (PNAs) are potential antisense therapies for genetic, acquired, and viral diseases. Efficiently selecting candidate PNA sequences for synthesis and evaluation from a genome containing hundreds to thousands of options can be challenging. To facilitate this process, this work leverages machine learning (ML) algorithms and automated synthesis technology to predict PNA synthesis efficiency and guide rational PNA sequence design. The training data is collected from individual fluorenylmethyloxycarbonyl (Fmoc) deprotection reactions performed on a fully automated PNA synthesizer. The optimized ML model allows for 93% prediction accuracy and 0.97 Pearson's r. The predicted synthesis scores are validated to be correlated with the experimental high-performance liquid chromatography (HPLC) crude purities (correlation coefficient R² = 0.95). Furthermore, a general applicability of ML is demonstrated through designing synthetically accessible antisense PNA sequences from 102 315 predicted candidates targeting exon 44 of the human dystrophin gene, SARS-CoV-2, HIV, as well as selected genes associated with cardiovascular diseases, type II diabetes, and various cancers. Collectively, ML provides an accurate prediction of PNA synthesis quality and serves as a useful computational tool for informing PNA sequence design.

Could artificial intelligence revolutionize the development of nanovectors for gene therapy and mRNA vaccines?

Gene therapy enables the introduction of nucleic acids like DNA and RNA into host cells, and is expected to revolutionize the treatment of a wide range of diseases. This growth has been further accelerated by the discovery of CRISPR/Cas technology, which allows accurate genomic editing in a broad range of cells and organisms in vitro and in vivo. Despite many advances in gene delivery and the development of various viral and non-viral gene delivery vectors, the lack of highly efficient non-viral systems with low cellular toxicity remains a challenge. The application of cutting-edge technologies such as artificial intelligence (AI) has great potential to find new paradigms to solve this issue. Herein, we review AI and its major subfields including machine learning (ML), neural networks (NNs), expert systems, deep learning (DL), computer vision and robotics. We discuss the potential of AI-based models and algorithms in the design of targeted gene delivery vehicles capable of crossing extracellular and intracellular barriers by viral mimicry strategies. We finally discuss the role of AI in improving the function of CRISPR/Cas systems, developing novel nanobots, and mRNA vaccine carriers.

Cell-penetrating peptides enhance peptide vaccine accumulation and persistence in lymph nodes to drive immunogenicity

Peptide-based cancer vaccines are widely investigated in the clinic but exhibit modest immunogenicity. One approach that has been explored to enhance peptide vaccine potency is covalent conjugation of antigens with cell-penetrating peptides (CPPs), linear cationic and amphiphilic peptide sequences designed to promote intracellular delivery of associated cargos. Antigen-CPPs have been reported to exhibit enhanced immunogenicity compared to free peptides, but their mechanisms of action in vivo are poorly understood. We tested eight previously described CPPs conjugated to antigens from multiple syngeneic murine tumor models and found that linkage to CPPs enhanced peptide vaccine potency in vivo by as much as 25-fold. Linkage of antigens to CPPs did not impact dendritic cell activation but did promote uptake of linked antigens by dendritic cells both in vitro and in vivo. However, T cell priming in vivo required Batf3-dependent dendritic cells, suggesting that antigens delivered by CPP peptides were predominantly presented via the process of cross-presentation and not through CPP-mediated cytosolic delivery of peptide to the classical MHC class I antigen processing pathway. Unexpectedly, we observed that many CPPs significantly enhanced antigen accumulation in draining lymph nodes. This effect was associated with the ability of CPPs to bind to lymph-trafficking lipoproteins and protection of CPP-antigens from proteolytic degradation in serum. These two effects resulted in prolonged presentation of CPP-peptides in draining lymph nodes, leading to robust T cell priming and expansion. Thus, CPPs can act through multiple unappreciated mechanisms to enhance T cell priming that can be exploited for cancer vaccines with enhanced potency.

Accurate Models of Substrate Preferences of Post-Translational Modification Enzymes from a Combination of mRNA Display and Deep Learning

Promiscuous post-translational modification (PTM) enzymes often display nonobvious substrate preferences by acting on diverse yet well-defined sets of peptides and/or proteins. Understanding of substrate fitness landscapes for PTM enzymes is important in many areas of contemporary science, including natural product biosynthesis, molecular biology, and biotechnology. Here, we report an integrated platform for accurate profiling of substrate preferences for PTM enzymes. The platform features (i) a combination of mRNA display with next-generation sequencing as an ultrahigh throughput technique for data acquisition and (ii) deep learning for data analysis. The high accuracy (>0.99 in each of two studies) of the resulting deep learning models enables comprehensive analysis of enzymatic substrate preferences. The models can quantify fitness across sequence space, map modification sites, and identify important amino acids in the substrate. To benchmark the platform, we performed profiling of a Ser dehydratase (LazBF) and a Cys/Ser cyclodehydratase (LazDEF), two enzymes from the lactazole biosynthesis pathway. In both studies, our results point to complex enzymatic preferences, which, particularly for LazBF, cannot be reduced to a set of simple rules. The ability of the constructed models to dissect such complexity suggests that the developed platform can facilitate a wider study of PTM enzymes.

Deep generative models for peptide design

Computers can already be programmed for superhuman pattern recognition of images and text. For machines to discover novel molecules, they must first be trained to sort through the many characteristics of molecules and determine which properties should be retained, suppressed, or enhanced to optimize functions of interest. Machines need to be able to understand, read, write, and eventually create new molecules. Today, this creative process relies on deep generative models, which have gained popularity since powerful deep neural networks were introduced to generative model frameworks. In recent years, they have demonstrated excellent ability to model complex distribution of real-word data (e.g., images, audio, text, molecules, and biological sequences). Deep generative models can generate data beyond those provided in training samples, thus yielding an efficient and rapid tool for exploring the massive search space of high-dimensional data such as DNA/protein sequences and facilitating the design of biomolecules with desired functions. Here, we review the emerging field of deep generative models applied to peptide science. In particular, we discuss several popular deep generative model frameworks as well as their applications to generate peptides with various kinds of properties (e.g., antimicrobial, anticancer, cell penetration, etc). We conclude our review with a discussion of current limitations and future perspectives in this emerging field.

Machine learning approaches for biomolecular, biophysical, and biomaterials research

A fluent conversation with a virtual assistant, person-tailored news feeds, and deep-fake images created within seconds-all those things that have been unthinkable for a long time are now a part of our everyday lives. What these examples have in common is that they are realized by different means of machine learning (ML), a technology that has fundamentally changed many aspects of the modern world. The possibility to process enormous amount of data in multi-hierarchical, digital constructs has paved the way not only for creating intelligent systems but also for obtaining surprising new insight into many scientific problems. However, in the different areas of biosciences, which typically rely heavily on the collection of time-consuming experimental data, applying ML methods is a bit more challenging: Here, difficulties can arise from small datasets and the inherent, broad variability, and complexity associated with studying biological objects and phenomena. In this Review, we give an overview of commonly used ML algorithms (which are often referred to as "machines") and learning strategies as well as their applications in different bio-disciplines such as molecular biology, drug development, biophysics, and biomaterials science. We highlight how selected research questions from those fields were successfully translated into machine readable formats, discuss typical problems that can arise in this context, and provide an overview of how to resolve those encountered difficulties.

Cell-Penetrating d-Peptides Retain Antisense Morpholino Oligomer Delivery Activity

Cell-penetrating peptides (CPPs) can cross the cell membrane to enter the cytosol and deliver otherwise nonpenetrant macromolecules such as proteins and oligonucleotides. For example, recent clinical trials have shown that a CPP attached to phosphorodiamidate morpholino oligomers (PMOs) resulted in higher muscle concentration, increased exon skipping, and dystrophin production relative to another study of the PMO alone in patients of Duchenne muscular dystrophy. Therefore, effective design and the study of CPPs could help enhance therapies for difficult-to-treat diseases. So far, the study of CPPs for PMO delivery has been restricted to predominantly canonical l-peptides. We hypothesized that mirror-image d-peptides could have similar PMO delivery activity as well as enhanced proteolytic stability, facilitating their characterization and quantification from biological milieu. We found that several enantiomeric peptide sequences could deliver a PMO-biotin cargo with similar activities while remaining stable against serum proteolysis. The biotin label allowed for affinity capture of fully intact PMO-peptide conjugates from whole-cell and cytosolic lysates. By profiling a mixture of these constructs in cells, we determined their relative intracellular concentrations. When combined with PMO activity, these concentrations provide a new metric for delivery efficiency, which may be useful for determining which peptide sequence to pursue in further preclinical studies.

Design of Protein Segments and Peptides for Binding to Protein Targets

Recent years have witnessed a rise in methods for accurate prediction of structure and design of novel functional proteins. Design of functional protein fragments and peptides occupy a small, albeit unique, space within the general field of protein design. While the smaller size of these peptides allows for more exhaustive computational methods, flexibility in their structure and sparsity of data compared to proteins, as well as presence of noncanonical building blocks, add additional challenges to their design. This review summarizes the current advances in the design of protein fragments and peptides for binding to targets and discusses the challenges in the field, with an eye toward future directions.

Automated Flow Synthesis of Peptide-PNA Conjugates

Antisense peptide nucleic acids (PNAs) have yet to translate to the clinic because of poor cellular uptake, limited solubility, and rapid elimination. Cell-penetrating peptides (CPPs) covalently attached to PNAs may facilitate clinical development by improving uptake into cells. We report an efficient technology that utilizes a fully automated fast-flow instrument to manufacture CPP-conjugated PNAs (PPNAs) in a single shot. The machine is rapid, with each amide bond being formed in 10 s. Anti-IVS2-654 PPNA synthesized with this instrument presented threefold activity compared to transfected PNA in a splice-correction assay. We demonstrated the utility of this approach by chemically synthesizing eight anti-SARS-CoV-2 PPNAs in 1 day. A PPNA targeting the 5' untranslated region of SARS-CoV-2 genomic RNA reduced the viral titer by over 95% in a live virus infection assay (IC₅₀ = 0.8 μM). Our technology can deliver PPNA candidates to further investigate their potential as antiviral agents.

Deep Learning Enables Discovery of a Short Nuclear Targeting Peptide for Efficient Delivery of Antisense Oligomers

Therapeutic macromolecules such as proteins and oligonucleotides can be highly efficacious but are often limited to extracellular targets due to the cell's impermeable membrane. Cell-penetrating peptides (CPPs) are able to deliver such macromolecules into cells, but limited structure-activity relationships and inconsistent literature reports make it difficult to design effective CPPs for a given cargo. For example, polyarginine motifs are common in CPPs, promoting cell uptake at the expense of systemic toxicity. Machine learning may be able to address this challenge by bridging gaps between experimental data in order to discern sequence-activity relationships that evade our intuition. Our earlier data set and deep learning model led to the design of miniproteins (>40 amino acids) for antisense delivery. Here, we leveraged and expanded our model with data augmentation in the short CPP sequence space of the data set to extrapolate and discover short, low-arginine-content CPPs that would be easier to synthesize and amenable to rapid conjugation to desired cargo, and with minimal in vivo toxicity. The lead predicted peptide, termed P6, is as active as a polyarginine CPP for the delivery of an antisense oligomer, while having only one arginine side chain and 18 total residues. We determined the pentalysine motif and the C-terminal cysteine of P6 to be the main drivers of activity. The antisense conjugate was able to enhance corrective splicing in an animal model to produce functional eGFP in heart tissue in vivo while remaining nontoxic up to a dose of 60 mg/kg. In addition, P6 was able to deliver an enzyme to the cytosol of cells. Our findings suggest that, given a data set of long CPPs, we can discover by extrapolation short, active sequences that deliver antisense oligomers.