WUREN: Whole-modal union representation for epitope prediction.

B-cell epitope identification plays a vital role in the development of vaccines, therapies, and diagnostic tools. Currently, molecular docking tools in B-cell epitope prediction are heavily influenced by empirical parameters and require significant computational resources, rendering a great challenge to meet large-scale prediction demands. When predicting epitopes from antigen-antibody complex, current artificial intelligence algorithms cannot accurately implement the prediction due to insufficient protein feature representations, indicating novel algorithm is desperately needed for efficient protein information extraction. In this paper, we introduce a multimodal model called WUREN (Whole-modal Union Representation for Epitope predictioN), which effectively combines sequence, graph, and structural features. It achieved AUC-PR scores of 0.213 and 0.193 on the solved structures and AlphaFold-generated structures, respectively, for the independent test proteins selected from DiscoTope3 benchmark. Our findings indicate that WUREN is an efficient feature extraction model for protein complexes, with the generalizable application potential in the development of protein-based drugs. Moreover, the streamlined framework of WUREN could be readily extended to model similar biomolecules, such as nucleic acids, carbohydrates, and lipids.

Au- ZnFe2O4 hollow microspheres based gas sensor for detecting the mustard gas simulant 2-chloroethyl ethyl sulfide.

Mustard gas, a representative of blister agents, poses a severe threat to human health. Although the structure of 2-chloroethyl ethyl sulfide (2-CEES) is similar to mustard gas, 2-CEES is non-toxic, rendering it a commonly employed simulant in related research. ZnFe2O4-based semiconductor gas sensors exhibit numerous advantages, including structural stability, high sensitivities, and easy miniaturization. However, they exhibit insufficient sensitivity at low concentrations and require high operating temperatures. Owing to the effect of electronic and chemical sensitization, the gas-sensing performance of a sensor may be remarkably enhanced via the sensitization method of noble metal loading. In this study, based on the morphologies of ZnFe2O4 hollow microspheres, a solvothermal method was adopted to realize different levels of Au loading. Toward 1 ppm of 2-CEES, the gas sensor based on 2 wt.% Au-loaded ZnFe2O4 hollow microspheres exhibited a response sensitivity twice that of the gas sensor based on pure ZnFe2O4; furthermore, the response/recovery times decreased. Additionally, the sensor displayed excellent linear response to low concentrations of 2-CEES, outstanding selectivity in the presence of several common volatile organic compounds, and good repeatability, as well as long-term stability. The Au-loaded ZnFe2O4-based sensor has considerable potential for use in detecting toxic chemical agents and their simulants.

Pd-Catalyzed Asymmetric Larock Indole Synthesis to Access Axially Chiral N-Arylindoles.

Larock indole synthesis is one of the most straightforward and efficient methods for the synthesis of indoles; however, there has been no asymmetric version yet for the construction of indole-based axially chiral N-arylindoles since its initial report in 1991. Herein we report the first example of an asymmetric Larock indole synthesis by employing a chiral sulfinamide phosphine (SadPhos) ligand (Ming-Phos) with palladium. It allows rapid construction of a wide range of axially chiral N-arylindole compounds in good yields up to 98:2 er. The application of this unique chiral scaffold as an organocatalyst is promising. Furthermore, a kinetic study has revealed that the alkyne migratory insertion is the rate-determining step, which has been proven by the density functional theory (DFT) calculations. Additionally, DFT studies also suggest that the N-C dihedral difference caused by the steric hindrance of the ligand contributes to enantioselectivity control.

Targeting plasmodium α-tubulin-1 to block malaria transmission to mosquitoes.

Plasmodium ookinetes use an invasive apparatus to invade mosquito midguts, and tubulins are the major structural proteins of this apical complex. We examined the role of tubulins in malaria transmission to mosquitoes. Our results demonstrate that the rabbit polyclonal antibodies (pAb) against human α-tubulin significantly reduced the number of P. falciparum oocysts in Anopheles gambiae midguts, while rabbit pAb against human ß-tubulin did not. Further studies showed that pAb, specifically against P. falciparum α-tubulin-1, also significantly limited P. falciparum transmission to mosquitoes. We also generated mouse monoclonal antibodies (mAb) using recombinant P. falciparum α-tubulin-1. Out of 16 mAb, two mAb, A3 and A16, blocked P. falciparum transmission with EC50 of 12 µg/ml and 2.8 µg/ml. The epitopes of A3 and A16 were determined to be a conformational and linear sequence of EAREDLAALEKDYEE, respectively. To understand the mechanism of the antibody-blocking activity, we studied the accessibility of live ookinete α-tubulin-1 to antibodies and its interaction with mosquito midgut proteins. Immunofluorescent assays showed that pAb could bind to the apical complex of live ookinetes. Moreover, both ELISA and pull-down assays demonstrated that insect cell-expressed mosquito midgut protein, fibrinogen-related protein 1 (FREP1), interacts with P. falciparum α-tubulin-1. Since ookinete invasion is directional, we conclude that the interaction between Anopheles FREP1 protein and Plasmodium α-tubulin-1 anchors and orients the ookinete invasive apparatus towards the midgut PM and promotes the efficient parasite infection in the mosquito.

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.

Development of a SAW poly(epichlorohydrin) gas sensor for detection of harmful chemicals.

The uniformity and compactness of the surface of a viscoelastic sensitive film are among the most important factors that influence the characteristics of a surface acoustic wave (SAW) gas sensor, directly affecting the detection sensitivity of a SAW sensor on a target gas. In this paper, poly(epichlorohydrin) (PECH) with viscoelastic properties was used as sensitive film for the detection of 2-chloroethyl ethyl sulfide (CEES), a common simulant of the chemical agent mustard gas. Nanoscale films were prepared using a spin coating technology on a SAW delay line of 200 MHz. Films were evaluated using polarizing microscopy and atomic force microscopy and observed with uniform surface states and particle diameter in the cluster region of 4.52-5.22 µm. The interface parameters, including contact angle, surface tension, Gibbs free energy, work of adhesion, work of immersion, and spreading coefficient values were 9.31° to 39.63°, 22.475 to 29.945 mN m-1, -85.70 to -78.08 J m-2, 78.08 to 85.70 J m-2, -42.62 to -35.00 J m-2, and 0.46 to 8.08 J m-1, respectively. These values were obtained by experiments combined with the Young T equation and Gibbs adsorption isotherm, and the surface analysis was carried out theoretically. The glass transition temperature (-22.4 °C), viscosity, pyrolysis, and other physical characteristics of the prepared PECH were discussed. Five SAW sensors prepared at the same time were used to test the repeatability of CEES measurements at one concentration, where the consistency of the sensor preparation was confirmed. At a concentration of 13.6 mg m-3 for CEES, 10 consecutive detection results showed good repeatability (i.e., standard deviation = 0.295, coefficient of variance = 0.021, and population mean deviation = 0.364). At room temperature (20 °C ± 5 °C), different concentrations of CEES were detected using the developed sensor, which showed good linearity in the concentration range of 1.9-19.6 mg m-3 (y = 0.0309 + 1.13x, r = 0.99478). The limit of detection was 0.85 mg m-3, the limit of quantitation was 1.91 mg m-3, and the sensitivity of the SAW sensor was 1.13 mV (mg m-3). The adsorption mechanism related to PECH in the detection of CEES was also discussed.

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.

Rapid de novo discovery of peptidomimetic affinity reagents for human angiotensin converting enzyme 2.

Rapid discovery and development of serum-stable, selective, and high affinity peptide-based binders to protein targets are challenging. Angiotensin converting enzyme 2 (ACE2) has recently been identified as a cardiovascular disease biomarker and the primary receptor utilized by the severe acute respiratory syndrome coronavirus 2. In this study, we report the discovery of high affinity peptidomimetic binders to ACE2 via affinity selection-mass spectrometry (AS-MS). Multiple high affinity ACE2-binding peptides (ABP) were identified by selection from canonical and noncanonical peptidomimetic libraries containing 200 million members (dissociation constant, KD = 19-123 nM). The most potent noncanonical ACE2 peptide binder, ABP N1 (KD = 19 nM), showed enhanced serum stability in comparison with the most potent canonical binder, ABP C7 (KD = 26 nM). Picomolar to low nanomolar ACE2 concentrations in human serum were detected selectively using ABP N1 in an enzyme-linked immunosorbent assay. The discovery of serum-stable noncanonical peptidomimetics like ABP N1 from a single-pass selection demonstrates the utility of advanced AS-MS for accelerated development of affinity reagents to protein targets.

Coulombic effects on resolution of ion mobility spectrometry and its application in online qualitative analysis.

Ion mobility spectrometry is an important gas analysis method used in the rapid detection field. However, due to a lacking of explicit mathematical model of ion peak, it is difficult to extract characteristic analyte peaks from a spectrum containing overlapping peaks to achieve online qualitative analysis. Here, we present an asymmetric peak model for processing ion mobility peaks. For the asymmetric peak model, the key is to accurately estimate the standard deviation of the peak model and the fitting function of the tailing edge. We focused on the Coulombic effects on resolution of ion mobility spectrometry based on a new hypothesis of ion cloud shape and derived a formula for calculating the standard deviation taking the initial pulse width, diffusion and Coulomb repulsion factors into account. The proposed asymmetric peak model combines the advantages of optimal physical and chemical interpretation and explicit mathematical meaning. A fast decomposition method based on the peak model was developed to decompose overlapping peaks. Two overlapping simulated data sets and one real data set (a mixture of acetone and methyl salicylate) were used to test the method. The results indicated that our proposed method successfully decomposed the overlapping spectrum into individual peaks and performed markedly better than other three available methods in terms of the execution time. The proposed method meets the requirements for online qualitative analysis.

An in vivo selection-derived d-peptide for engineering erythrocyte-binding antigens that promote immune tolerance.

When displayed on erythrocytes, peptides and proteins can drive antigen-specific immune tolerance. Here, we investigated a straightforward approach based on erythrocyte binding to promote antigen-specific tolerance to both peptides and proteins. We first identified a robust erythrocyte-binding ligand. A pool of one million fully d-chiral peptides was injected into mice, blood cells were isolated, and ligands enriched on these cells were identified using nano-liquid chromatography-tandem mass spectrometry. One round of selection yielded a murine erythrocyte-binding ligand with an 80 nM apparent dissociation constant, Kd We modified an 83-kDa bacterial protein and a peptide antigen derived from ovalbumin (OVA) with the identified erythrocyte-binding ligand. An administration of the engineered bacterial protein led to decreased protein-specific antibodies in mice. Similarly, mice given the engineered OVA-derived peptide had decreased inflammatory anti-OVA CD8+ T cell responses. These findings suggest that our tolerance-induction strategy is applicable to both peptide and protein antigens and that our in vivo selection strategy can be used for de novo discovery of robust erythrocyte-binding ligands.

Automated affinity selection for rapid discovery of peptide binders.

In-solution affinity selection (AS) of large synthetic peptide libraries affords identification of binders to protein targets through access to an expanded chemical space. Standard affinity selection methods, however, can be time-consuming, low-throughput, or provide hits that display low selectivity to the target. Here we report an automated bio-layer interferometry (BLI)-assisted affinity selection platform. When coupled with tandem mass spectrometry (MS), this method enables both rapid de novo discovery and affinity maturation of known peptide binders with high selectivity. The BLI-assisted AS-MS technology also features real-time monitoring of the peptide binding during the library selection process, a feature unattainable by current selection approaches. We show the utility of the BLI AS-MS platform toward rapid identification of novel nanomolar (dissociation constant, K D < 50 nM) non-canonical binders to the leukemia-associated oncogenic protein menin. To our knowledge, this is the first application of BLI to the affinity selection of synthetic peptide libraries. We believe our approach can significantly accelerate the use of synthetic peptidomimetic libraries in drug discovery.

Fully automated fast-flow synthesis of antisense phosphorodiamidate morpholino oligomers.

Rapid development of antisense therapies can enable on-demand responses to new viral pathogens and make personalized medicine for genetic diseases practical. Antisense phosphorodiamidate morpholino oligomers (PMOs) are promising candidates to fill such a role, but their challenging synthesis limits their widespread application. To rapidly prototype potential PMO drug candidates, we report a fully automated flow-based oligonucleotide synthesizer. Our optimized synthesis platform reduces coupling times by up to 22-fold compared to previously reported methods. We demonstrate the power of our automated technology with the synthesis of milligram quantities of three candidate therapeutic PMO sequences for an unserved class of Duchenne muscular dystrophy (DMD). To further test our platform, we synthesize a PMO that targets the genomic mRNA of SARS-CoV-2 and demonstrate its antiviral effects. This platform could find broad application not only in designing new SARS-CoV-2 and DMD antisense therapeutics, but also for rapid development of PMO candidates to treat new and emerging diseases.

De Novo Discovery of High-Affinity Peptide Binders for the SARS-CoV-2 Spike Protein.

The ß-coronavirus SARS-CoV-2 has caused a global pandemic. Affinity reagents targeting the SARS-CoV-2 spike protein are of interest for the development of therapeutics and diagnostics. We used affinity selection-mass spectrometry for the rapid discovery of synthetic high-affinity peptide binders for the receptor binding domain (RBD) of the SARS-CoV-2 spike protein. From library screening with 800 million synthetic peptides, we identified three sequences with nanomolar affinities (dissociation constants K d = 80-970 nM) for RBD and selectivity over human serum proteins. Nanomolar RBD concentrations in a biological matrix could be detected using the biotinylated lead peptide in ELISA format. These peptides do not compete for ACE2 binding, and their site of interaction on the SARS-CoV-2-spike-RBD might be unrelated to the ACE2 binding site, making them potential orthogonal reagents for sandwich immunoassays. These findings serve as a starting point for the development of SARS-CoV-2 diagnostics or conjugates for virus-directed delivery of therapeutics.

Discovery of Nucleic Acid Binding Molecules from Combinatorial Biohybrid Nucleobase Peptide Libraries.

Nature has three biopolymers: oligonucleotides, polypeptides, and oligosaccharides. Each biopolymer has independent functions, but when needed, they form mixed assemblies for higher-order purposes, as in the case of ribosomal protein synthesis. Rather than forming large complexes to coordinate the role of different biopolymers, we dovetail protein amino acids and nucleobases into a single low molecular weight precision polyamide polymer. We established efficient chemical synthesis and de novo sequencing procedures and prepared combinatorial libraries with up to 100 million biohybrid molecules. This biohybrid material has a higher bulk affinity to oligonucleotides than peptides composed exclusively of canonical amino acids. Using affinity selection mass spectrometry, we discovered variants with a high affinity for pre-microRNA hairpins. Our platform points toward the development of high throughput discovery of sequence defined polymers with designer properties, such as oligonucleotide binding.

Selectively Suppressing Tumor Angiogenesis for Targeted Breast Cancer Therapy by Genetically Engineered Phage.

Antiangiogenesis is a promising approach to cancer therapy but is limited by the lack of tumor-homing capability of the current antiangiogenic agents. Angiogenin, a protein overexpressed and secreted by tumors to trigger angiogenesis for their growth, has never been explored as an antiangiogenic target in cancer therapy. Here it is shown that filamentous fd phage, as a biomolecular biocompatible nanofiber, can be engineered to become capable of first homing to orthotopic breast tumors and then capturing angiogenin to prevent tumor angiogenesis, resulting in targeted cancer therapy without side effects. The phage is genetically engineered to display many copies of an identified angiogenin-binding peptide on its side wall and multiple copies of a breast-tumor-homing peptide at its tip. Since the tumor-homing peptide can be discovered and customized virtually toward any specific cancer by phage display, the angiogenin-binding phages are thus universal "plug-and-play" tumor-homing cancer therapeutics.

Multiscale orthogonal matching pursuit algorithm combined with peak model for interpreting ion mobility spectra and achieving quantitative analysis.

Ion mobility spectrometry is an important rapid analysis method. However, it is difficult to achieve quantitative analysis when spectral peaks overlap. A new method for analyzing ion mobility spectra is presented here. The method achieves quantitative analysis by combining the advantages of the peak model (in terms of optimal physical and chemical interpretation of the system of interest) and the multiscale orthogonal matching pursuit algorithm (in terms of extracting characteristic peaks). A simulated data set, constructed using the peak model, containing overlapping peaks was analyzed to demonstrate the ability of the multiscale orthogonal matching pursuit algorithm to decompose overlapping peaks. Real data sets for methyl salicylate and a mixture of acetone and methyl salicylate at sixteen concentrations were generated using a vapor generator (using permeation tubes). The characteristic peaks were extracted using the multiscale orthogonal matching pursuit algorithm. Univariate calibrations using the peak area and peak height were prepared to allow quantitative analyses to be performed. Multivariate calibrations using partial-least-squares and poly-partial-least-squares were prepared and the results were compared with the univariate calibration results. Markedly better or similar predictions were made using the univariate calibration models involving physical and chemical interpretations than using the multivariate calibration models.

Towards early monitoring of chemotherapy-induced drug resistance based on single cell metabolomics: Combining single-probe mass spectrometry with machine learning.

Despite the presence of methods evaluating drug resistance during chemotherapies, techniques, which allow for monitoring the degree of drug resistance in early chemotherapeutic stage from single cells in their native microenvironment, are still absent. Herein, we report an analytical approach that combines single cell mass spectrometry (SCMS) based metabolomics with machine learning (ML) models to address the existing challenges. Metabolomic profiles of live cancer cells (HCT-116) with different levels (i.e., no, low, and high) of chemotherapy-induced drug resistance were measured using the Single-probe SCMS technique. A series of ML models, including random forest (RF), artificial neural network (ANN), and penalized logistic regression (LR), were constructed to predict the degrees of drug resistance of individual cells. A systematic comparison of performance was conducted among multiple models, and the method validation was carried out experimentally. Our results indicate that these ML models, especially the RF model constructed on the obtained SCMS datasets, can rapidly and accurately predict different degrees of drug resistance of live single cells. With such rapid and reliable assessment of drug resistance demonstrated at the single cell level, our method can be potentially employed to evaluate chemotherapeutic efficacy in the clinic.

Design, Synthesis, and Evaluation of 18F-Labeled Monoacylglycerol Lipase Inhibitors as Novel Positron Emission Tomography Probes.

Dysfunction of monoacylglycerol lipase (MAGL) is associated with several psychopathological disorders, including drug addiction and neurodegenerative diseases. Herein we design, synthesize, and evaluate several irreversible fluorine-containing MAGL inhibitors for positron emission tomography (PET) ligand development. Compound 6 (identified from a therapeutic agent) was advanced for 18F-labeling via a novel spirocyclic iodonium ylide (SCIDY) strategy, which demonstrated high brain permeability and excellent specific binding. This work supports further development of novel 18F-labeled MAGL PET probes.

A fast progressive spectrum denoising combined with partial least squares algorithm and its application in online Fourier transform infrared quantitative analysis.

Fourier transform infrared (FTIR) spectroscopy is an important method in analytical chemistry. A material can be qualitatively and quantitatively analyzed from its FTIR spectrum. Spectrum denoising is commonly performed before online FTIR quantitative analysis. The average method requires a long time to collect spectra, which weakens real-time online analysis. The Savitzky-Golay smoothing method makes peaks smoother with the increase of window width, causing useful information to be lost. The sparse representation method is a common denoising method, that is used to reconstruct spectrum. However, for the randomness of noise, we can't achieve the sparse representation of noise. Traditional sparse representation algorithms only perform denoising once, and the noise can not be removed completely. FTIR spectrum denoising should therefore be performed in a progressive way. However, it is difficult to determine to what degree of denoising is required. Here, a fast progressive spectrum denoising combined with partial least squares method was developed for online FTIR quantitative analysis. Two real sample data sets were used to test the performance of the proposed method. The experimental results indicated that the progressive spectrum denoising method combined with the partial least squares method performed markedly better than other methods in terms of root mean squared error of prediction and coefficient of determination in the FTIR quantitative analysis.

Anticancer Drug Affects Metabolomic Profiles in Multicellular Spheroids: Studies Using Mass Spectrometry Imaging Combined with Machine Learning.

Multicellular spheroids (hereinafter referred to as spheroids) are 3D biological models. The metabolomic profiles inside spheroids provide crucial information reflecting the molecular phenotypes and microenvironment of cells. To study the influence of an anticancer drug on the spatially resolved metabolites, spheroids were cultured using HCT-116 colorectal cancer cells, treated with the anticancer drug Irinotecan under a series of time- and concentration-dependent conditions. The Single-probe mass spectrometry imaging (MSI) technique was utilized to conduct the experiments. The MSI data were analyzed using advanced data analysis methods to efficiently extract metabolomic information. Multivariate curve resolution alternating least square (MCR-ALS) was used to decompose each MS image into different components with grouped species. To improve the efficiency of data analysis, both supervised (Random Forest) and unsupervised (cluster large applications (CLARA)) machine learning (ML) methods were employed to cluster MS images according to their metabolomic features. Our results indicate that anticancer drug significantly affected the abundances of a variety of metabolites in different regions of spheroids. This integrated experiment and data analysis approach can facilitate the studies of metabolites in different types of 3D tumor models and tissues and potentially benefit the drug discovery, therapeutic resistance, and other biological research fields.