Sequence Design Research Articles

Accurate and efficient protein sequence design through learning concise local environment of residues.

Computational protein sequence design has been widely applied in rational protein engineering and increasing the design accuracy and efficiency is highly desired. Here, we present ProDESIGN-LE, an accurate and efficient approach to protein sequence design. ProDESIGN-LE adopts a concise but informative representation of the residue's local environment and trains a transformer to learn the correlation between local environment of residues and their amino acid types. For a target backbone structure, ProDESIGN-LE uses the transformer to assign an appropriate residue type for each position based on its local environment within this structure, eventually acquiring a designed sequence with all residues fitting well with their local environments. We applied ProDESIGN-LE to design sequences for 68 naturally occurring and 129 hallucinated proteins within 20 s per protein on average. The designed proteins have their predicted structures perfectly resembling the target structures with a state-of-the-art average TM-score exceeding 0.80. We further experimentally validated ProDESIGN-LE by designing five sequences for an enzyme, chloramphenicol O-acetyltransferase type III (CAT III), and recombinantly expressing the proteins in Escherichia coli. Of these proteins, three exhibited excellent solubility, and one yielded monomeric species with circular dichroism spectra consistent with the natural CAT III protein. The source code of ProDESIGN-LE is available at https://github.com/bigict/ProDESIGN-LE.

Community science designed ribosomes with beneficial phenotypes

Functional design of ribosomes with mutant ribosomal RNA (rRNA) can expand opportunities for understanding molecular translation, building cells from the bottom-up, and engineering ribosomes with altered capabilities. However, such efforts are hampered by cell viability constraints, an enormous combinatorial sequence space, and limitations on large-scale, 3D design of RNA structures and functions. To address these challenges, we develop an integrated community science and experimental screening approach for rational design of ribosomes. This approach couples Eterna, an online video game that crowdsources RNA sequence design to community scientists in the form of puzzles, with in vitro ribosome synthesis, assembly, and translation in multiple design-build-test-learn cycles. We apply our framework to discover mutant rRNA sequences that improve protein synthesis in vitro and cell growth in vivo, relative to wild type ribosomes, under diverse environmental conditions. This work provides insights into rRNA sequence-function relationships and has implications for synthetic biology.

Network of hotspot interactions cluster tau amyloid folds

Cryogenic electron microscopy has revealed unprecedented molecular insight into the conformations of β-sheet-rich protein amyloids linked to neurodegenerative diseases. It remains unknown how a protein can adopt a diversity of folds and form multiple distinct fibrillar structures. Here we develop an in silico alanine scan method to estimate the relative energetic contribution of each amino acid in an amyloid assembly. We apply our method to twenty-seven ex vivo and in vitro fibril structural polymorphs of the microtubule-associated protein tau. We uncover networks of energetically important interactions involving amyloid-forming motifs that stabilize the different fibril folds. We evaluate our predictions in cellular and in vitro aggregation assays. Using a machine learning approach, we classify the structures based on residue energetics to identify distinguishing and unifying features. Our energetic profiling suggests that minimal sequence elements control the stability of tau fibrils, allowing future design of protein sequences that fold into unique structures.

Copolymers with Nonblocky Sequences as Novel Materials with Finely Tuned Properties

The copolymer sequence can be considered as a new tool to shape the resulting system properties on demand. This perspective is devoted to copolymers with "partially segregated" (or nonblocky) sequences. Such copolymers include gradient copolymers and copolymers with random sequences as well as copolymers with precisely controlled sequences. We overview recent developments in the synthesis of these systems as well as new findings regarding their properties, in particular, self-assembly in solutions and in melts. An emphasis is put on how the microscopic behavior of polymer chains is influenced by the chain sequences. In addition to that, a novel class of approaches allowing one to efficiently tackle the problem of copolymer chain sequence design─data driven methods (artificial intelligence and machine learning)─is discussed.

Influence of drilling sequence and guide-hole design on the accuracy of static computer-assisted implant surgery in extraction sockets and healed sites-An in vitro investigation.

To evaluate the effect of drilling sequence, guide-hole design, and alveolar ridge morphology on the accuracy of implant placement via static computer-assisted implant surgery (sCAIS). Standardized maxillary bone models including single-tooth gaps with fresh extraction sockets or healed alveolar ridge morphologies were evaluated in this study. Implants were placed using different drilling sequences (i.e., complete [CDS] or minimum [MDS]), and guide-hole designs (i.e., manufacturer's sleeve [MS] or sleeveless [SL] guide-hole designs). The time for implant placement via sCAIS procedures was also recorded. The angular, crestal, and apical three-dimensional deviations between planned and final implant positions were digitally obtained. Statistical analyses were conducted by a non-parametric three-way ANOVA (α= .05). Based on a sample size analysis, a total of 72 implants were included in this study. Significantly higher implant position accuracy was found at healed sites compared to extraction sockets and in SL compared to MS guide-hole design in angular, crestal, and apical 3D deviations (p≤ .048). A tendency for higher accuracy was observed for the CDS compared to the MDS, although the effect was not statistically significant (p= .09). The MDS required significantly shorter preparation times compared with CDS (p< .0001). Implant placement via sCAIS resulted in higher accuracy in healed sites than extraction sockets, when using SL compared to MS guides, and tended to be more accurate when using CDS compared to MDS. Therefore, even though surgery time was shorter with MDS, its use should be limited to strictly selected cases.

Novel insights into construct toxicity, strain optimization, and primary sequence design for producing recombinant silk fibroin and elastin-like peptide in E. coli

Spider silk proteins (spidroins) are a remarkable class of biomaterials that exhibit a unique combination of high-value attributes and can be processed into numerous morphologies for targeted applications in diverse fields. Recombinant production of spidroins represents the most promising route towards establishing the industrial production of the material, however, recombinant spider silk production suffers from fundamental difficulties that includes low titers, plasmid instability, and translational inefficiencies. In this work, we sought to gain a deeper understanding of upstream bottlenecks that exist in the field through the production of a panel of systematically varied spidroin sequences in multiple E. coli strains. A restriction on basal expression and specific genetic mutations related to stress responses were identified as primary factors that facilitated higher titers of the recombinant silk constructs. Using these findings, a novel strain of E. coli was created that produces recombinant silk constructs at levels 4–33 times higher than standard BL21(DE3). However, these findings did not extend to a similar recombinant protein, an elastin-like peptide. It was found that the recombinant silk proteins, but not the elastin-like peptide, exert toxicity on the E. coli host system, possibly through their high degree of intrinsic disorder. Along with strain engineering, a bioprocess design that utilizes longer culturing times and attenuated induction was found to raise recombinant silk titers by seven-fold and mitigate toxicity. Targeted alteration to the primary sequence of the recombinant silk constructs was also found to mitigate toxicity. These findings identify multiple points of focus for future work seeking to further optimize the recombinant production of silk proteins and is the first work to identify the intrinsic disorder and subsequent toxicity of certain spidroin constructs as a primary factor related to the difficulties of production.

An evaluation of force field accuracy for the mini-protein chignolin using Markov state models.

All-atom molecular dynamics (MD) simulations can provide detailed insight into a molecule's conformational ensemble in solution. While molecular force fields are parameterized to accurately model a protein's potential energy surface, it remains challenging in practice to evaluate how well force fields can capture ensemble-averaged experimental observables, since it requires simulation of the complete folding landscape. In this work, we employ massively parallel molecular simulations, performed using the [email protected] distributed computing platform, to investigate the ability of nine force fields (AMBER14SB, AMBER99, AMBER99SB, AMBER99SB-ildn, AMBER99SBnmr1-ildn, CHARMM22∗, CHARMM27, CHARMM36 and OPLS-aa) with TIP3P explicit solvent to accurately reproduce experimental observables for chignolin, a beta-hairpin mini-protein with an experimental folding time of ∼600 ns. From over 200 microseconds of aggregate trajectory data, we constructed Markov state models (MSMs) to obtain estimates of thermodynamic and kinetic properties of chignolin in each force field. Quantitative assessment of the force fields was performed by comparing predicted and experimental folded populations, and the statistical agreement between predicted and experimental solution-state NMR observables. Our results rank force field accuracy for chignolin similarly compared to previous work using oligopeptides and ubiquitin (Beauchamp et al. 2012). This work highlights the utility of MSM approaches for force field evaluation, and provides a baseline for future studies using Bayesian inference methods to evaluate and parameterize force fields. It also provides an excellent dataset for testing new approaches to simulation-based sequence design of mini-proteins.

Comparative analysis of kinetic typography elements in movie title sequence

Typography, usually closely related to printed media, has now been explored in digital media. These changes make typography, which is usually static, more dynamic. One example of digital typography is kinetic typography. Typography's position as a medium of visual and textual communication becomes more interesting because the element of motion in kinetic typography adds another meaning value. This study examines how typography reveals meaning in time-based media formats. In this study, three movie title sequences with different genres will be analyzed: Monsters Inc., Se7en, and Mission: Impossible — Ghost Protocol. The film was chosen because it has a dominant kinetic typography structure and has a good reputation in terms of visuals. The method of analysis is carried out by comparing the structure of the three title sequences based on the kinetic typography theory by Bellantoni and Woolman. Furthermore, an analysis of the semantic value of each title sequence was also carried out. This research is aimed to figure out the elements used in the visual creation of a title sequence and how these elements can construct meaning. From the analysis process it is known that each genre of the title sequence has different element variations which of course have their own meaning. The results of this study are expected to provide recommendations for the title sequence design method based on the genre like comedy animation, thriller, and action.

Scalable Combinatorial Assembly of Synthetic DNA for Tracking Applications.

Synthetic DNA barcodes are double-stranded DNA molecules designed to carry recoverable information, information that can be used to represent and track objects and organisms. DNA barcodes offer robust, sensitive detection using standard amplification and sequencing techniques. While numerous research groups have promoted DNA as an information storage medium, less attention has been devoted to the design of economical, scalable DNA barcode libraries. Here, we present an alternative modular approach to sequence design. Barcode sequences were constructed from smaller, interchangeable blocks, allowing for the combinatorial assembly of numerous distinct tags. We demonstrated the design and construction of first-generation (N = 256) and second-generation (N = 512) modular barcode libraries, from fewer than 50 total single-stranded oligonucleotides for each library. To avoid contamination during experimental validation, a liquid-handling robot was employed for oligonucleotide mixing. Generating barcode sequences in-house reduces dependency upon external entities for unique tag generation, increasing flexibility in barcode generation and deployment. Next generation sequencing (NGS) detection of 256 different samples in parallel highlights the multiplexing afforded by the modular barcode design coupled with high-throughput sequencing. Deletion variant analysis of the first-generation library informed sequence design for enhancing barcode assembly specificity in the second-generation library.

Flexible Cyclic Immunofluorescence (cyCIF) Using Oligonucleotide Barcoded Antibodies.

Advances in our understanding of the complex, multifaceted interactions between tumor epithelia, immune infiltrate, and tumor microenvironmental cells have been driven by highly multiplexed imaging technologies. These techniques are capable of labeling many more biomarkers than conventional immunostaining methods. However, multiplexed imaging techniques suffer from low detection sensitivity, cell loss-particularly in fragile samples-, and challenges with antibody labeling. Herein, we developed and optimized an oligonucleotide antibody barcoding strategy for cyclic immunofluorescence (cyCIF) that can be amplified to increase the detection efficiency of low-abundance antigens. Stained fluorescence signals can be readily removed using ultraviolet light treatment, preserving tissue and fragile cell sample integrity. We also extended the oligonucleotide barcoding strategy to secondary antibodies to enable the inclusion of difficult-to-label primary antibodies in a cyCIF panel. Using both the amplification oligonucleotides to label DNA barcoded antibodies and in situ hybridization of multiple fluorescently labeled oligonucleotides resulted in signal amplification and increased signal-to-background ratios. This procedure was optimized through the examination of staining parameters including staining oligonucleotide concentration, staining temperature, and oligonucleotide sequence design, resulting in a robust amplification technique. As a proof-of-concept, we demonstrate the flexibility of our cyCIF strategy by simultaneously imaging with the original oligonucleotide conjugated antibody (Ab-oligo) cyCIF strategy, the novel Ab-oligo cyCIF amplification strategy, as well as direct and indirect immunofluorescence to generate highly multiplexed images.

3D RNA-scaffolded wireframe origami

Hybrid RNA:DNA origami, in which a long RNA scaffold strand folds into a target nanostructure via thermal annealing with complementary DNA oligos, has only been explored to a limited extent despite its unique potential for biomedical delivery of mRNA, tertiary structure characterization of long RNAs, and fabrication of artificial ribozymes. Here, we investigate design principles of three-dimensional wireframe RNA-scaffolded origami rendered as polyhedra composed of dual-duplex edges. We computationally design, fabricate, and characterize tetrahedra folded from an EGFP-encoding messenger RNA and de Bruijn sequences, an octahedron folded with M13 transcript RNA, and an octahedron and pentagonal bipyramids folded with 23S ribosomal RNA, demonstrating the ability to make diverse polyhedral shapes with distinct structural and functional RNA scaffolds. We characterize secondary and tertiary structures using dimethyl sulfate mutational profiling and cryo-electron microscopy, revealing insight into both global and local, base-level structures of origami. Our top-down sequence design strategy enables the use of long RNAs as functional scaffolds for complex wireframe origami.

Benchmarking TriadAb using targets from the second antibody modeling assessment.

Computational modeling and design of antibodies has become an integral part of today's research and development in antibody therapeutics. Here we describe the Triad Antibody Homology Modeling (TriadAb) package, a functionality of the Triad protein design platform that predicts the structure of any heavy and light chain sequences of an antibody Fv domain using template-based modeling. To gauge the performance of TriadAb, we benchmarked against the results of the Second Antibody Modeling Assessment (AMA-II). On average, TriadAb produced main-chain carbonyl root-mean-square deviations between models and experimentally determined structures at 1.10Å, 1.45Å, 1.41Å, 3.04Å, 1.47Å, 1.27Å, 1.63Å in the framework and the six complementarity-determining regions (H1, H2, H3, L1, L2, L3), respectively. The inaugural results are comparable to those reported in AMA-II, corroborating with our internal bench-based experiences that models generated using TriadAb are sufficiently accurate and useful for antibody engineering using the sequence design capabilities provided by Triad.

Neural network-derived Potts models for structure-based protein design using backbone atomic coordinates and tertiary motifs.

Designing novel proteins to perform desired functions, such as binding or catalysis, is a major goal in synthetic biology. A variety of computational approaches can aid in this task. An energy-based framework rooted in the sequence-structure statistics of tertiary motifs (TERMs) can be used for sequence design on predefined backbones. Neural network models that use backbone coordinate-derived features provide another way to design new proteins. In this work, we combine the two methods to make neural structure-based models more suitable for protein design. Specifically, we supplement backbone-coordinate features with TERM-derived data, as inputs, and we generate energy functions as outputs. We present two architectures that generate Potts models over the sequence space: TERMinator, which uses both TERM-based and coordinate-based information, and COORDinator, which uses only coordinate-based information. Using these two models, we demonstrate that TERMs can be utilized to improve native sequence recovery performance of neural models. Furthermore, we demonstrate that sequences designed by TERMinator are predicted to fold to their target structures by AlphaFold. Finally, we show that both TERMinator and COORDinator learn notions of energetics, and these methods can be fine-tuned on experimental data to improve predictions. Our results suggest that using TERM-based and coordinate-based features together may be beneficial for protein design and that structure-based neural models that produce Potts energy tables have utility for flexible applications in protein science.

Improved Multi-Strategy Matrix Particle Swarm Optimization for DNA Sequence Design

The efficiency of DNA computation is closely related to the design of DNA coding sequences. For the purpose of obtaining superior DNA coding sequences, it is necessary to choose suitable DNA constraints to prevent potential conflicting interactions in different DNA sequences and to ensure the reliability of DNA sequences. An improved matrix particle swarm optimization algorithm, referred to as IMPSO, is proposed in this paper to optimize DNA sequence design. In addition, this paper incorporates centroid opposition-based learning to fully preserve population diversity and develops and adapts a dynamic update on the basis of signal-to-noise ratio distance to search for high-quality solutions in a sufficiently intelligent manner. The results show that the proposal of this paper achieves satisfactory results and can obtain higher computational efficiency.

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.

Automated Multi- and Block-Copolymer Writing in Mesoporous Films Using Visible-Light PET-RAFT and a Microscope.

For high throughput applications, e.g., in the context of sensing especially when being combined with machine learning, large sample numbers in acceptable production time are required. This needs automated synthesis and material functionalization concepts ideally combined with high precision. To automate sensing relevant mesopore polymer functionalization while being highly precise in polymer placement, polymer amount control, and polymer sequence design, a process for polymer writing in mesoporous silica films with pore diameter in the range of 13nm is developed. Mesoporous films are functionalized with different polymers in adjustable polymer amount including block-copolymer functionalization in an automated process using a visible-light induced, controlled photo electron/energy transfer-reversible addition-fragmentation chain-transfer (PET-RAFT) polymerization. While transferring this PET-RAFT to a commercially available microscope, direct, automated laser writing of three different polymers, as well as polymer re-initiation is demonstrated. Using a laser diameter of ≈72µm, significantly smaller polymer spots of ≈7µm in diameter are realized. Micrometerscale resolved polymer images including block-copolymers are written into mesoporous layers covering millimeter scale areas requiring a writing time in the range of one second per polymer spot.

Sequence Design of Random Heteropolymers as Protein Mimics.

Random heteropolymers (RHPs) have been computationally designed and experimentally shown to recapitulate protein-like phase behavior and function. However, unlike proteins, RHP sequences are only statistically defined and cannot be sequenced. Recent developments in reversible-deactivation radical polymerization allowed simulated polymer sequences based on the well-established Mayo-Lewis equation to more accurately reflect ground-truth sequences that are experimentally synthesized. This led to opportunities to perform bioinformatics-inspired analysis on simulated sequences to guide the design, synthesis, and interpretation of RHPs. We compared batches on the order of 10000 simulated RHP sequences that vary by synthetically controllable and measurable RHP characteristics such as chemical heterogeneity and average degree of polymerization. Our analysis spans across 3 levels: segments along a single chain, sequences within a batch, and batch-averaged statistics. We discuss simulator fidelity and highlight the importance of robust segment definition. Examples are presented that demonstrate the use of simulated sequence analysis for in-silico iterative design to mimic protein hydrophobic/hydrophilic segment distributions in RHPs and compare RHP and protein sequence segments to explain experimental results of RHPs that mimic protein function. To facilitate the community use of this workflow, the simulator and analysis modules have been made available through an open source toolkit, the RHPapp.

Reconfigurable Peptide Analogs of Apolipoprotein A-I Reveal Tunable Features of Nanodisc Assembly.

Nanodisc (ND)-forming membrane scaffold proteins or peptides developed from apolipoprotein A-I (apoA-I) have led to considerable promise in structural biology and therapeutic applications. However, the rationale and regularity characteristics in peptide sequence design remain inconclusive. Here, we proposed a consensus-based normalization approach through the reversed engineering of apoA-IΔ1-45 to design reconfigurable apoA-I peptide analogs (APAs) for tunable ND assembly. We present extensive morphological validations and computational simulation analyses on divergent APA-NDs that are generated by our method. Fifteen divergent APAs were generated accordingly to study the assembly machinery of NDs. We show that APA designs exhibit multifactorial influence in terms of varying APA tandem repeats, sequence composition, and lipid-to-APA ratio to form tunable diameters of NDs. There is a strong positive correlation between DMPC-to-APA ratios and ND diameters. Longer APA with more tandem repeats tends to yield higher particle size homogeneity. Our results also suggest proline is a dispensable residue for the APA-ND formation. Interestingly, proline-rich substitution not only provides an inward-bending effect in forming smaller NDs but also induces the cumulative chain flexibility that enables larger ND formation at higher lipid ratios. Additionally, proline-tryptophan residues in APAs play a dominant role in forming larger NDs. Molecular simulation shows that enriched basic and acidic residues in APAs evoke abundant hydrogen-bond and salt bridge networks to reinforce the structural stability of APA-NDs. Together, our findings provide a rational basis for understanding APA design. The proposed model could be extended to other apolipoproteins for desired ND engineering.

Effective and efficient gradient based methods for low-bit discrete-phase sequence designs

We devise effective and efficient gradient based methods to design low-bit, including binary (one-bit per sample, or 1b) and quaternary (two-bit per sample, or 2b), discrete-phase sequences without or with low correlation zone (LCZ) requirements. We determine proper step sizes needed by the gradient based algorithms to monotonically decrease the optimization metric via efficient line search techniques for aperiodic and periodic low-bit discrete-phase sequence designs. Moreover, we introduce approximate gradient based approaches to reduce the autocorrelation sidelobes when the desired LCZ is relatively large. Numerical examples are provided to show that the proposed gradient based algorithms can outperform the state-of-the-art algorithms for designing 1b and 2b sequences with good autocorrelation properties.

Ultrasensitive CRISPR/Cas13a-Mediated Photoelectrochemical Biosensors for Specific and Direct Assay of miRNA-21.

Sensitive and specific assay of microRNAs (miRNAs) is beneficial to early disease screening. Herein, we for the first time proposed clustered regularly interspaced short palindromic repeats (CRISPR)/Cas13a-mediated photoelectrochemical biosensors for the direct assay of miRNA-21. In this study, compared with traditional nucleic acid-based signal amplification strategies, the CRISPR/Cas13a system can greatly improve the specificity and sensitivity of target determination due to its accurate recognition and high-efficient trans-cleavage capability without complex nucleic acid sequence design. Moreover, compared with the CRISPR/Cas12a-based biosensing platform, the developed CRISPR/Cas13a-mediated biosensor can directly detect RNA targets without signal transduction from RNA to DNA, thereby avoiding signal leakage and distortion. Generally, the proposed biosensor reveals excellent analysis capability with a wider linear range from 1 fM to 5 nM and a lower detection limit of 1 fM. Additionally, it also shows satisfactory stability in the detection of human serum samples and cell lysates, manifesting that it has great application prospects in the areas of early disease diagnosis and biomedical research.