Expanding the Cell-Free Reporter Protein Toolbox by Employing a Split mNeonGreen System to Reduce Protein Synthesis Workload.

The cell-free system offers potential advantages in biosensor applications, but its limited time for protein synthesis poses a challenge in creating enough fluorescent signals to detect low limits of the analyte while providing a robust sensing module at the beginning. In this study, we harnessed split versions of fluorescent proteins, particularly split superfolder green fluorescent protein and mNeonGreen, to increase the number of reporter units made before the reaction ceased and enhance the detection limit in the cell-free system. A comparative analysis of the expression of 1-10 and 11th segments of beta strands in both whole-cell and cell-free platforms revealed distinct fluorescence patterns. Moreover, the integration of SynZip peptide linkers substantially improved complementation. The split protein reporter system could enable higher reporter output when sensing low analyte levels in the cell-free system, broadening the toolbox of the cell-free biosensor repertoire.

Deconvoluting complex correlates of COVID-19 severity with a multi-omic pandemic tracking strategy.

The SARS-CoV-2 pandemic has differentially impacted populations across race and ethnicity. A multi-omic approach represents a powerful tool to examine risk across multi-ancestry genomes. We leverage a pandemic tracking strategy in which we sequence viral and host genomes and transcriptomes from nasopharyngeal swabs of 1049 individuals (736 SARS-CoV-2 positive and 313 SARS-CoV-2 negative) and integrate them with digital phenotypes from electronic health records from a diverse catchment area in Northern California. Genome-wide association disaggregated by admixture mapping reveals novel COVID-19-severity-associated regions containing previously reported markers of neurologic, pulmonary and viral disease susceptibility. Phylodynamic tracking of consensus viral genomes reveals no association with disease severity or inferred ancestry. Summary data from multiomic investigation reveals metagenomic and HLA associations with severe COVID-19. The wealth of data available from residual nasopharyngeal swabs in combination with clinical data abstracted automatically at scale highlights a powerful strategy for pandemic tracking, and reveals distinct epidemiologic, genetic, and biological associations for those at the highest risk.

Characterizing a New Fluorescent Protein for a Low Limit of Detection Sensing in the Cell-Free System.

Cell-free protein synthesis-based biosensors have been developed as highly accurate, low-cost biosensors. However, since most biomarkers exist at low concentrations in various types of biopsies, the biosensor's dynamic range must be increased in the system to achieve low limits of detection necessary while deciphering from higher background signals. Many attempts to increase the dynamic range have relied on amplifying the input signal from the analyte, which can lead to complications of false positives. In this study, we aimed to increase the protein synthesis capability of the cell-free protein synthesis system and the output signal of the reporter protein to achieve a lower limit of detection. We utilized a new fluorescent protein, mNeonGreen, which produces a higher output than those commonly used in cell-free biosensors. Optimizations of DNA sequence and the subsequent cell-free protein synthesis reaction conditions allowed characterizing protein expression variability by given DNA template types, reaction environment, and storage additives that cause the greatest time constraint on designing the cell-free biosensor. Finally, we characterized the fluorescence kinetics of mNeonGreen compared to the commonly used reporter protein, superfolder green fluorescent protein. We expect that this finely tuned cell-free protein synthesis platform with the new reporter protein can be used with sophisticated synthetic gene circuitry networks to increase the dynamic range of a cell-free biosensor to reach lower detection limits and reduce the false-positive proportion.

Effect of Acid Mixtures on Surface Properties and Biaxial Flexural Strength of As-Sintered and Air-Abraded Zirconia.

The aim of this work was to evaluate the effects of application time of an acid mixture solution on the surface roughness, phase transformation, and biaxial flexural strength of 3Y-TZP after sintering or air abrasion. For the biaxial flexural strength measurement, 220 3Y-TZP disk-shaped specimens were prepared after as-sintering or air abrasion. The etching solution comprised a mixture of hydrofluoric acid, sulfuric acid, hydrogen peroxide, methyl alcohol, and purified water. The samples were divided into 11 subgroups according to the etching times (Control, 1, 2, 3, 5, 8, 10, 12, 15, 20, and 30 min). The results showed that acid treatment on both as-sintered and air-abraded 3Y-TZP surfaces increased the surface roughness. However, it had no significant effects on the monoclinic phase or flexural strength of as-sintered zirconia. The monoclinic phase and flexural strength of air-abraded zirconia increased sharply after air abrasion; however, they gradually decreased after acid treatment, to a similar level to the case of the untreated surface. Surface treatment with acid mixture increased the roughness, but the lack of increase of monoclinic phase is thought to be because the loose monoclinic particles remaining on the surface were removed through the etching process.

Tuning the Cell-Free Protein Synthesis System for Biomanufacturing of Monomeric Human Filaggrin.

The modern cell-free protein synthesis (CFPS) system is expanding the opportunity of cell-free biomanufacturing as a versatile platform for synthesizing various therapeutic proteins. However, synthesizing human protein in the bacterial CFPS system remains challenging due to the low expression level, protein misfolding, inactivity, and more. These challenges limit the use of a bacterial CFPS system for human therapeutic protein synthesis. In this study, we demonstrated the improved performance of a customized CFPS platform for human therapeutic protein production by investigating the factors that limit cell-free transcription-translation. The improvement of the CFPS platform has been made in three ways. First, the cell extract was prepared from the rare tRNA expressed host strain, and CFPS was performed with a codon-optimized gene for Escherichia coli codon usage bias. The soluble protein yield was 15.2 times greater with the rare tRNA overexpressing host strain as cell extract and codon-optimized gene in the CFPS system. Next, we identify and prioritize the critical biomanufacturing factors for highly active crude cell lysate for human protein synthesis. Lastly, we engineer the CFPS reaction conditions to enhance protein yield. In this model, the therapeutic protein filaggrin expression was significantly improved by up to 23-fold, presenting 28 ± 5 µM of soluble protein yield. The customized CFPS system for filaggrin biomanufacturing described here demonstrates the potential of the CFPS system to be adapted for studying therapeutic proteins.

Comprehensive Study on Molecular Supervised Learning with Graph Neural Networks.

This work considers strategies to develop accurate and reliable graph neural networks (GNNs) for molecular property predictions. Prediction performance of GNNs is highly sensitive to the change in various parameters due to the inherent challenges in molecular machine learning, such as a deficient amount of data samples and bias in data distribution. Comparative studies with well-designed experiments are thus important to clearly understand which GNNs are powerful for molecular supervised learning. Our work presents a number of ablation studies along with a guideline to train and utilize GNNs for both molecular regression and classification tasks. First, we validate that using both atomic and bond meta-information improves the prediction performance in the regression task. Second, we find that the graph isomorphism hypothesis proposed by [Xu, K.; et al How powerful are graph neural networks? 2018, arXiv:1810.00826. arXiv.org e-Print archive. https://arxiv.org/abs/1810.00826] is valid for the regression task. Surprisingly, however, the findings above do not hold for the classification tasks. Beyond the study on model architectures, we test various regularization methods and Bayesian learning algorithms to find the best strategy to achieve a reliable classification system. We demonstrate that regularization methods penalizing predictive entropy might not give well-calibrated probability estimation, even though they work well in other domains, and Bayesian learning methods are capable of developing reliable prediction systems. Furthermore, we argue the importance of Bayesian learning in virtual screening by showing that well-calibrated probability estimation may lead to a higher success rate.

A Surgical Simulator for Tympanostomy Tube Insertion Incorporating Capacitive Sensing Technology to Track Instrument Placement.

We describe a device engineered for realistic simulation of myringotomy and tympanostomy tube insertion that tracks instrument placement and objectively measures operator proficiency. A 3-dimensional computer model of the external ear and cartilaginous external auditory canal was created from a normal maxillofacial computed tomography scan, and models for the bony external auditory canal and tympanic cavity were created with computer-aided design software. Physical models were 3-dimensionally printed from the computer reconstructions. The external auditory canal and tympanic cavity surfaces were coated with conductive material and wired to a capacitive sensor interface. A programmable microcontroller with custom embedded software completed the system. Construct validation was completed by comparing the run times and total sensor contact times of otolaryngology faculty and residents.

A Bayesian graph convolutional network for reliable prediction of molecular properties with uncertainty quantification.

Deep neural networks have been increasingly used in various chemical fields. In the nature of a data-driven approach, their performance strongly depends on data used in training. Therefore, models developed in data-deficient situations can cause highly uncertain predictions, leading to vulnerable decision making. Here, we show that Bayesian inference enables more reliable prediction with quantitative uncertainty analysis. Decomposition of the predictive uncertainty into model- and data-driven uncertainties allows us to elucidate the source of errors for further improvements. For molecular applications, we devised a Bayesian graph convolutional network (GCN) and evaluated its performance for molecular property predictions. Our study on the classification problem of bio-activity and toxicity shows that the confidence of prediction can be quantified in terms of the predictive uncertainty, leading to more accurate virtual screening of drug candidates than standard GCNs. The result of log P prediction illustrates that data noise affects the data-driven uncertainty more significantly than the model-driven one. Based on this finding, we could identify artefacts that arose from quantum mechanical calculations in the Harvard Clean Energy Project dataset. Consequently, the Bayesian GCN is critical for molecular applications under data-deficient conditions.

A Crude Extract Preparation and Optimization from a Genomically Engineered Escherichia coli for the Cell-Free Protein Synthesis System: Practical Laboratory Guideline.

With the advancement of synthetic biology, the cell-free protein synthesis (CFPS) system has been receiving the spotlight as a versatile toolkit for engineering natural and unnatural biological systems. The CFPS system reassembles the materials necessary for transcription and translation and recreates the in vitro protein synthesis environment by escaping a physical living boundary. The cell extract plays an essential role in this in vitro format. Here, we propose a practical protocol and method for Escherichia coli-derived cell extract preparation and optimization, which can be easily applied to both commercially available and genomically engineered E. coli strains. The protocol includes: (1) The preparation step for cell growth and harvest, (2) the thorough step-by-step procedures for E. coli cell extract preparation including the cell wash and lysis, centrifugation, runoff reaction, and dialysis, (3) the preparation for the CFPS reaction components and, (4) the quantification of cell extract and cell-free synthesized protein. We anticipate that the protocol in this research will provide a simple preparation and optimization procedure of a highly active E. coli cell extract.

ISLES 2016 and 2017-Benchmarking Ischemic Stroke Lesion Outcome Prediction Based on Multispectral MRI.

Performance of models highly depend not only on the used algorithm but also the data set it was applied to. This makes the comparison of newly developed tools to previously published approaches difficult. Either researchers need to implement others' algorithms first, to establish an adequate benchmark on their data, or a direct comparison of new and old techniques is infeasible. The Ischemic Stroke Lesion Segmentation (ISLES) challenge, which has ran now consecutively for 3 years, aims to address this problem of comparability. ISLES 2016 and 2017 focused on lesion outcome prediction after ischemic stroke: By providing a uniformly pre-processed data set, researchers from all over the world could apply their algorithm directly. A total of nine teams participated in ISLES 2015, and 15 teams participated in ISLES 2016. Their performance was evaluated in a fair and transparent way to identify the state-of-the-art among all submissions. Top ranked teams almost always employed deep learning tools, which were predominately convolutional neural networks (CNNs). Despite the great efforts, lesion outcome prediction persists challenging. The annotated data set remains publicly available and new approaches can be compared directly via the online evaluation system, serving as a continuing benchmark (www.isles-challenge.org).

Cell-free protein synthesis from genomically recoded bacteria enables multisite incorporation of noncanonical amino acids.

Cell-free protein synthesis has emerged as a powerful approach for expanding the range of genetically encoded chemistry into proteins. Unfortunately, efforts to site-specifically incorporate multiple non-canonical amino acids into proteins using crude extract-based cell-free systems have been limited by release factor 1 competition. Here we address this limitation by establishing a bacterial cell-free protein synthesis platform based on genomically recoded Escherichia coli lacking release factor 1. This platform was developed by exploiting multiplex genome engineering to enhance extract performance by functionally inactivating negative effectors. Our most productive cell extracts enabled synthesis of 1,780 ± 30 mg/L superfolder green fluorescent protein. Using an optimized platform, we demonstrated the ability to introduce 40 identical p-acetyl-L-phenylalanine residues site specifically into an elastin-like polypeptide with high accuracy of incorporation ( ≥ 98%) and yield (96 ± 3 mg/L). We expect this cell-free platform to facilitate fundamental understanding and enable manufacturing paradigms for proteins with new and diverse chemistries.

A Pressure Test to Make 10 Molecules in 90 Days: External Evaluation of Methods to Engineer Biology.

Centralized facilities for genetic engineering, or "biofoundries", offer the potential to design organisms to address emerging needs in medicine, agriculture, industry, and defense. The field has seen rapid advances in technology, but it is difficult to gauge current capabilities or identify gaps across projects. To this end, our foundry was assessed via a timed "pressure test", in which 3 months were given to build organisms to produce 10 molecules unknown to us in advance. By applying a diversity of new approaches, we produced the desired molecule or a closely related one for six out of 10 targets during the performance period and made advances toward production of the others as well. Specifically, we increased the titers of 1-hexadecanol, pyrrolnitrin, and pacidamycin D, found novel routes to the enediyne warhead underlying powerful antimicrobials, established a cell-free system for monoterpene production, produced an intermediate toward vincristine biosynthesis, and encoded 7802 individually retrievable pathways to 540 bisindoles in a DNA pool. Pathways to tetrahydrofuran and barbamide were designed and constructed, but toxicity or analytical tools inhibited further progress. In sum, we constructed 1.2 Mb DNA, built 215 strains spanning five species ( Saccharomyces cerevisiae, Escherichia coli, Streptomyces albidoflavus, Streptomyces coelicolor, and Streptomyces albovinaceus), established two cell-free systems, and performed 690 assays developed in-house for the molecules.

Rapid production and characterization of antimicrobial colicins using Escherichia coli-based cell-free protein synthesis.

Colicins are antimicrobial proteins produced by Escherichia coli, which, upon secretion from the host, kill non-host E. coli strains by forming pores in the inner membrane and degrading internal cellular components such as DNA and RNA. Due to their unique cell-killing activities, colicins are considered viable alternatives to conventional antibiotics. Recombinant production of colicins requires co-production of immunity proteins to protect host cells; otherwise, the colicins are lethal to the host. In this study, we used cell-free protein synthesis (CFPS) to produce active colicins without the need for protein purification and co-production of immunity proteins. Cell-free synthesized colicins were active in killing model E. coli cells with different modes of cytotoxicity. Pore-forming colicins E1 and nuclease colicin E2 killed actively growing cells in a nutrient-rich medium, but the cytotoxicity of colicin Ia was low compared to E1 and E2. Moreover, colicin E1 effectively killed cells in a nutrient-free solution, while the activity of E2 was decreased compared to nutrient-rich conditions. Both colicins E1 and E2 decreased the level of persister cells (metabolically dormant cell populations that are insensitive to antibiotics) by up to six orders of magnitude compared to that of the rifampin pretreated persister cells. This study finds that colicins can eradicate non-growing cells including persisters, and that CFPS is a promising platform for rapid production and characterization of toxic proteins.

Establishing a high yielding streptomyces-based cell-free protein synthesis system.

Cell-free protein synthesis (CFPS) has emerged as a powerful platform for applied biotechnology and synthetic biology, with a range of applications in synthesizing proteins, evolving proteins, and prototyping genetic circuits. To expand the current CFPS repertoire, we report here the development and optimization of a Streptomyces-based CFPS system for the expression of GC-rich genes. By developing a streamlined crude extract preparation protocol and optimizing reaction conditions, we were able to achieve active enhanced green fluorescent protein (EGFP) yields of greater than 50 µg/mL with batch reactions lasting up to 3 h. By adopting a semi-continuous reaction format, the EGFP yield could be increased to 282 ± 8 µg/mL and the reaction time was extended to 48 h. Notably, our extract preparation procedures were robust to multiple Streptomyces lividans and Streptomyces coelicolor strains, although expression yields varied. We show that our optimized Streptomyces lividans system provides benefits when compared to an Escherichia coli-based CFPS system for increasing percent soluble protein expression for four Streptomyces-originated high GC-content genes that are involved in biosynthesis of the nonribosomal peptides tambromycin and valinomycin. Looking forward, we believe that our Streptomyces-based CFPS system will contribute significantly towards efforts to express complex natural product gene clusters (e.g., nonribosomal peptides and polyketides), providing a new avenue for obtaining and studying natural product biosynthesis pathways. Biotechnol. Bioeng. 2017;114: 1343-1353. © 2017 Wiley Periodicals, Inc.

Calibrated propensity score method for survey nonresponse in cluster sampling.

Weighting adjustment is commonly used in survey sampling to correct for unit nonresponse. In cluster sampling, the missingness indicators are often correlated within clusters and the response mechanism is subject to cluster-specific nonignorable missingness. Based on a parametric working model for the response mechanism that incorporates cluster-specific nonignorable missingness, we propose a method of weighting adjustment. We provide a consistent estimator of the mean or totals in cases where the study variable follows a generalized linear mixed-effects model. The proposed method is robust in the sense that the consistency of the estimator does not require correct specification of the functional forms of the response and outcome models. A consistent variance estimator based on Taylor linearization is also proposed. Numerical results, including a simulation and a real-data application, are presented.

High-throughput preparation methods of crude extract for robust cell-free protein synthesis.

Crude extract based cell-free protein synthesis (CFPS) has emerged as a powerful technology platform for high-throughput protein production and genetic part characterization. Unfortunately, robust preparation of highly active extracts generally requires specialized and costly equipment and can be labor and time intensive. Moreover, cell lysis procedures can be hard to standardize, leading to different extract performance across laboratories. These challenges limit new entrants to the field and new applications, such as comprehensive genome engineering programs to improve extract performance. To address these challenges, we developed a generalizable and easily accessible high-throughput crude extract preparation method for CFPS based on sonication. To validate our approach, we investigated two Escherichia coli strains: BL21 Star™ (DE3) and a K12 MG1655 variant, achieving similar productivity (defined as CFPS yield in g/L) by varying only a few parameters. In addition, we observed identical productivity of cell extracts generated from culture volumes spanning three orders of magnitude (10 mL culture tubes to 10 L fermentation). We anticipate that our rapid and robust extract preparation method will speed-up screening of genomically engineered strains for CFPS applications, make possible highly active extracts from non-model organisms, and promote a more general use of CFPS in synthetic biology and biotechnology.

Improving cell-free protein synthesis through genome engineering of Escherichia coli lacking release factor 1.

Site-specific incorporation of non-standard amino acids (NSAAs) into proteins opens the way to novel biological insights and applications in biotechnology. Here, we describe the development of a high yielding cell-free protein synthesis (CFPS) platform for NSAA incorporation from crude extracts of genomically recoded Escherichia coli lacking release factor 1. We used genome engineering to construct synthetic organisms that, upon cell lysis, lead to improved extract performance. We targeted five potential negative effectors to be disabled: the nuclease genes rna, rnb, csdA, mazF, and endA. Using our most productive extract from strain MCJ.559 (csdA(-) endA(-)), we synthesized 550±40 µg mL(-1) of modified superfolder green fluorescent protein containing p-acetyl-L-phenylalanine. This yield was increased to ∼1300 µg mL(-1) when using a semicontinuous method. Our work has implications for using whole genome editing for CFPS strain development, expanding the chemistry of biological systems, and cell-free synthetic biology.

Non-standard amino acid incorporation into proteins using Escherichia coli cell-free protein synthesis.

Incorporating non-standard amino acids (NSAAs) into proteins enables new chemical properties, new structures, and new functions. In recent years, improvements in cell-free protein synthesis (CFPS) systems have opened the way to accurate and efficient incorporation of NSAAs into proteins. The driving force behind this development has been three-fold. First, a technical renaissance has enabled high-yielding (>1 g/L) and long-lasting (>10 h in batch operation) CFPS in systems derived from Escherichia coli. Second, the efficiency of orthogonal translation systems (OTSs) has improved. Third, the open nature of the CFPS platform has brought about an unprecedented level of control and freedom of design. Here, we review recent developments in CFPS platforms designed to precisely incorporate NSAAs. In the coming years, we anticipate that CFPS systems will impact efforts to elucidate structure/function relationships of proteins and to make biomaterials and sequence-defined biopolymers for medical and industrial applications.

Cloning-independent expression and screening of enzymes using cell-free protein synthesis systems.

We present a strategy for expression and screening of microbial enzymes without involving cloning procedures. Libraries of putative ω-transaminases (ω-TA) and mutated Candida antarctica lipase B (CalB) are PCR-amplified from bacterial colonies and directly expressed in an Escherichia coli-based cell-free protein synthesis system. The open nature of cell-free protein synthesis system also allows streamlined analysis of the enzymatic activity of the expressed enzymes, which greatly shortens the time required for enzyme screening. We expect that the proposed strategy will provide a universal platform for bridging the information gap between nucleotide sequence and protein function, in order to accelerate the discovery of novel enzymes. The proposed strategy can also serve as a viable option for the rapid and precise tuning of enzyme molecules, not only for analytical purposes, but also for industrial applications. This is accomplished via large-scale production using microbial cells transformed with variant genes selected from the cell-free expression screening.