Search

Proteolytic targeting chimera (PROTAC) represent an advanced strategy for targeting undruggable proteins, and the molecular warheads targeting E3 ligases play a crucial role. Recently, we explored an alkenyl oxindole warhead targeting the E3 ligase DCAF11 and sought to validate its potential. In this study, we synthesized a range of BRD4 PROTACs ( 8a - 8o , 14a-14f , 22a-22m ) with modified alkenyl oxindole warheads and developed a high-throughput screening system based on high-content imaging. We identified L134 ( 22a ) as a potent BRD4 degrader, achieving BRD4 degradation ( D max > 98%, DC 50 = 7.36 nM) and demonstrating antitumor activity. Mechanically, BRD4 degradation by L134 was mediated through the ubiquitin-proteasome system in a DCAF11-dependent manner. Therefore, this study provides a rapid screening method for effective PROTACs and highlights the PROTAC L134 based on alkenyl oxindole-DCAF11 pair as a promising candidate for treating BRD4-driven cancers.

Technological advances and breakthrough developments in the pharmaceutical field are knocking at the door of the "undruggable" fortress with increasing insistence. Notably, the 21st century has seen the emergence of macrocyclic compounds, among which cyclic peptides are of particular interest. This new class of potential drug candidates occupies the vast chemical space between classic small-molecule drugs and larger protein-based therapeutics, such as antibodies. As research advances toward clinical targets that have long been considered inaccessible, macrocyclic peptides are well-suited to tackle these challenges in a post-rule of 5 pharmaceutical landscape. Facilitating their discovery is an arsenal of high-throughput screening methods that exploit massive randomized libraries of genetically encoded compounds. These techniques benefit from the incorporation of non-natural moieties, such as non- proteinogenic amino acids or stabilizing hydrocarbon staples. Exploiting these features for the strategic architectural design of macrocyclic peptides has the potential to tackle challenging targets such as protein-protein interactions, which have long resisted research efforts. This Review summarizes the basic principles and recent developments of the main high-throughput techniques for the discovery of macrocyclic peptides and focuses on their specific deployment for targeting undruggable space. A particular focus is placed on the development of new design guidelines and principles for the cyclization and structural stabilization of cyclic peptides and the resulting success stories achieved against well-known inaccessible drug targets.

The accurate and effective determination of antimicrobial resistance is essential to limiting the spread of infectious diseases and ensuring human health. Herein, a simple, accurate, and high-throughput phage-based colorimetric sensing strategy was developed for antimicrobial susceptibility testing (AST). Taking advantage of the CRISPR/Cas9 system, the genome of the T4 phage was modularly engineered to carry lacZ-α (lacZa), a marker gene encoding the α-fragment of β-galactosidase (β-gal). T4 lacZa phages were identified by blue-white selection and then used for a biosensing application. In this strategy, the bacterial solution is exposed to the T4 lacZa phage, causing target bacteria to overexpress β-gal. Upon the addition of a colorimetric substrate, the β-gal initiates an enzymatic reaction, resulting in a solution color change from yellow to red. This sensing strategy offers a visual way to monitor bacterial growth in the presence of antibiotics, enabling the determination of bacterial antimicrobial susceptibility. As a proof of concept, our developed sensing strategy was successfully applied to identify 9 different multidrug-resistant Escherichia coli ( E. coli ) in urine samples with 100% specificity. Compared with conventional disk diffusion susceptibility tests, the engineered phage-based sensing strategy can shorten the detection time by at least half without losing detection sensitivity, providing an alternative high-throughput method for AST in clinical diagnosis.

COVID-19 made apparent the devastating impact viral pandemics have had on global health and order. Development of broad-spectrum antivirals to provide early protection upon the inevitable emergence of new viral pandemics is critical. In this work, antiviral polymers are discovered using a combination of high-throughput polymer synthesis and antiviral screening, enabling diverse polymer compositions to be explored. Amphipathic polymers, with ionizable tertiary amine groups, are the most potent antivirals, effective against influenza virus and SARS-CoV-2, with minimal cytotoxicity. It is hypothesized that these polymers interact with the viral membrane as they showed no activity against a nonenveloped virus (rhinovirus). The switchable chemistry of the polymers during endosomal acidification was evaluated using lipid monolayers, indicating that a complex synergy between hydrophobicity and ionization drives polymer-membrane interactions. This new high-throughput methodology can be adapted to continue to engineer the potency of the lead candidates or develop antiviral polymers against other emerging viral classes.

Current developments in single-cell mass spectrometry (MS) aim to deepen proteome coverage while enhancing analytical speed to study entire cell populations, one cell at a time. Custom-built microanalytical capillary electrophoresis (μCE) played a critical role in the foundation of discovery single-cell MS proteomics. However, requirements for manual operation, substantial expertise, and low measurement throughput have so far hindered μCE-based single-cell studies on large numbers of cells. Here, we design and construct a robotic capillary (RoboCap) platform that grants single-cell CE-MS with automation for proteomes limited to less than ∼100 nL. RoboCap remotely controls precision actuators to translate the sample to the fused silica separation capillary, using vials in this work. The platform is hermetically enclosed and actively pressurized to inject ∼1-250 nL of the sample into a CE separation capillary, with errors below ∼5% relative standard deviation (RSD). The platform and supporting equipment were operated and monitored remotely on a custom-written Virtual Instrument (LabView). Detection performance was validated empirically on ∼5-250 nL portions of the HeLa proteome digest using a trapped ion mobility mass spectrometer (timsTOF PRO). RoboCap improved CE-ESI sample utilization to ∼20% from ∼3% on the manual μCE, the closest reference technology. Proof-of-principle experiments found proteome identification and quantification to robustly return ∼1,800 proteins (∼13% RSD) from ∼20 ng of the HeLa proteome digest on this earlier-generation detector. RoboCap automates CE-MS for limited sample amounts, paving the way to electrophoresis-based high-throughput single-cell proteomics.

Nonspecific reactive chemicals often interfere with the interpretation of high-throughput assay results because of their promiscuity and/or cytotoxicity. Using a high-throughput assay to identify such compounds is necessary to efficiently rule out potential assay artifacts. The MSTI, ( E )-2-(4-mercaptostyryl)-1,3,3-trimethyl-3 H -indol-1-ium, assay uses a thiol-containing fluorescent probe to screen for electrophile reactivity and could potentially be used to determine nonspecific reactive compounds. The Tox21 10K compound library was previously screened against a panel of ∼80 cell-based and biochemical assays, including the biochemical MSTI assay. In this study, we compared the MSTI assay activity of the Tox21 10K compounds with their promiscuity and cytotoxicity as reflected by their activities across the Tox21 assay panel to determine: (1) if this assay is predictive of a compound's promiscuity and cytotoxicity and (2) what chemical features create inconsistent results between the MSTI assay activity and promiscuity/cytotoxicity (false negatives and false positives). We found that the MSTI assay can predict a chemical's promiscuity/cytotoxicity with a 0.55 sensitivity and 0.97 specificity. Out of 3,407 unique compounds evaluated, we identified 92 false positive and 227 false negative results. Several structural features such as carboxamides and alkyl halides were found to be apparent in 53% ( p = 2.4 × 10 -07 ) and 19% ( p = 4.3 × 10 -06 ) of the false positives and negatives, respectively. The results of this analysis will help identify the potential challenges of this high-throughput assay and allow researchers to identify if a compound will be cytotoxic or promiscuous in an efficient manner.

Metabolic abnormalities play a pivotal role in various pathological conditions, necessitating the quantification of specific metabolites for diagnosis. While mass spectrometry remains the primary method for metabolite measurement, its limited throughput underscores the need for biosensors capable of rapid detection. Previously, we reported that pillar[6]arene with 12 carboxylate groups (P6AC) forms host-guest complexes with 1-methylnicotinamide (1-MNA), which is produced in vivo by nicotinamide N -methyltransferase (NNMT). P6AC acts as a biosensor by measuring the fluorescence quenching caused by photoinduced electron transfer upon 1-MNA binding. However, the low sensitivity of P6AC makes it impractical for detecting 1-MNA in unpurified biological samples. In this study, we found that P6A with 12 sulfonate groups (P6AS) is a specific and potent supramolecular host for 1-MNA interactions even in biological samples. The 1-MNA binding affinity of P6AS in water was found to be (5.68 ± 1.02) × 10 6 M -1 , which is approximately 700-fold higher than that of P6AC. Moreover, the 1-MNA detection limit of P6AS was determined to be 2.84 × 10 -7 M, which is substantially lower than that of P6AC. Direct addition of P6AS to culture medium was sufficient to quantify 1-MNA produced by cancer cells. Furthermore, this sensor was able to specifically detect 1-MNA even in unpurified human urine. P6AS therefore enables rapid and high-throughput quantification of 1-MNA, and further improvement of our strategy will contribute to the establishment of high-throughput screening of NNMT inhibitors, diagnosis of liver diseases, and imaging of human cancer cells in vivo.

Nucleophilic index ( N Nu ) as a significant parameter plays a crucial role in screening of amine catalysts. Indeed, the quantity and variety of amines are extensive. However, only limited amines exhibit an N Nu value exceeding 4.0 eV, rendering them potential nucleophiles in chemical reactions. To address this issue, we proposed a computational method to quickly identify amines with high N Nu values by using Machine Learning (ML) and high-throughput Density Functional Theory (DFT) calculations. Our approach commenced by training ML models and the exploration of Molecular Fingerprint methods as well as the development of quantitative structure-activity relationship (QSAR) models for the well-known amines based on N Nu values derived from DFT calculations. Utilizing explainable Shapley Additive Explanation plots, we were able to determine the five critical substructures that significantly impact the N Nu values of amine. The aforementioned conclusion can be applied to produce and cultivate 4920 novel hypothetical amines with high N Nu values. The QSAR models were employed to predict the N Nu values of 259 well-known and 4920 hypothetical amines, resulting in the identification of five novel hypothetical amines with exceptional N Nu values (>4.55 eV). The enhanced N Nu values of these novel amines were validated by DFT calculations. One novel hypothetical amine, H1, exhibits an unprecedentedly high N Nu value of 5.36 eV, surpassing the maximum value (5.35 eV) observed in well-established amines. Our research strategy efficiently accelerates the discovery of the high nucleophilicity of amines using ML predictions, as well as the DFT calculations.

The enzyme glutamate-cysteine ligase catalytic subunit (Gclc) is a rate-limiting enzyme in the biosynthesis of glutathione that is involved in antioxidant defense, detoxification of xenobiotics, and/or its metabolites and regulates the cell cycle and immune function. Therefore, Gclc presents an appealing target for the development of novel insecticides. In this study, we conducted high-throughput virtual screening from the ZINC20 database and identified three compounds with high binding affinity to the Tribolium castaneum Gclc (TcGclc). Ultimately, we selected ZINC000032992384 due to its superior stability and lowest binding energy, as determined through molecular dynamics simulations. Bioassay results revealed that the IC 50 value of ZINC000032992384 was 19.70 μM lower than that of BSO (49.67 μM). Furthermore, the larval mortality in the ZINC000032992384 treated group was 63.8%, significantly higher than that of the controls (29.1% in the dichlorvos group and 6.4% in the acetone group). This study provides novel insights for the development of a Gclc-targeted inhibitor as a potent insecticide based on the interaction between receptors and ligands.

Monitoring and quantifying host cell proteins (HCPs) in biotherapeutic production processes is crucial to ensure product quality, stability, and safety. Liquid chromatography-mass spectrometry (LC-MS) analysis has emerged as an important tool for identifying and quantifying individual HCPs. However, LC-MS-based approaches face challenges due to the wide dynamic range between HCPs and the therapeutic protein as well as laborious sample preparation and long instrument time. To address these limitations, we evaluated the application of parallel accumulation-serial fragmentation combined with data-independent acquisition (diaPASEF) to HCP analysis for biopharmaceutical process development applications. We evaluated different library generation strategies and LC methods, demonstrating the suitability of these workflows for various HCP analysis needs, such as in-depth characterization and high-throughput analysis of process intermediates. Remarkably, the diaPASEF approach enabled the quantification of hundreds of HCPs that were undetectable by a standard data-dependent acquisition mode while considerably improving sample requirement, throughput, coverage, quantitative precision, and data completeness.

Liquid chromatography-mass spectrometry (LC-MS) based metabolomics suffers from extended duty cycles and matrix-dependent quantitation. Chemical tags with 96 unique masses are reported, which alleviate the metabolomic workflow bottleneck and allow for absolute quantitation. A metabolic screen for carboxylic acids was performed on mammalian cells deprived of various nutrients and showed 24% RSD and analysis of 288 samples in 2 h.

Monoacylglycerol lipase (MAGL) is the key enzyme for the hydrolysis of endocannabinoid 2-arachidonoylglycerol (2-AG). The central role of MAGL in the metabolism of 2-AG makes it an attractive therapeutic target for a variety of disorders, including inflammation-induced tissue injury, pain, multiple sclerosis, and cancer. Previously, we reported LEI-515 , an aryl sulfoxide, as a peripherally restricted, covalent reversible MAGL inhibitor that reduced neuropathic pain and inflammation in preclinical models. Here, we describe the structure-activity relationship (SAR) of aryl sulfoxides as MAGL inhibitors that led to the identification of LEI-515 . Optimization of the potency of high-throughput screening (HTS) hit 1 yielded compound ±43 . However, ±43 was not metabolically stable due to its ester moiety. Replacing the ester group with α-CF 2 ketone led to the identification of compound ±73 ( LEI-515 ) as a metabolically stable MAGL inhibitor with subnanomolar potency. LEI-515 is a promising compound to harness the therapeutic potential of MAGL inhibition.

Glutaric Aciduria Type 1 (GA1) is a serious inborn error of metabolism with no pharmacological treatments. A novel strategy to treat this disease is to divert the toxic biochemical intermediates to less toxic or nontoxic metabolites. Here, we report a putative novel target, succinyl-CoA:glutarate-CoA transferase (SUGCT), which we hypothesize suppresses the GA1 metabolic phenotype through decreasing glutaryl-CoA and the derived 3-hydroxyglutaric acid. SUGCT is a type III CoA transferase that uses succinyl-CoA and glutaric acid as substrates. We report the structure of SUGCT, develop enzyme- and cell-based assays, and identify valsartan and losartan carboxylic acid as inhibitors of the enzyme in a high-throughput screen of FDA-approved compounds. The cocrystal structure of SUGCT with losartan carboxylic acid revealed a novel pocket in the active site and further validated the high-throughput screening approach. These results may form the basis for the future development of new pharmacological intervention to treat GA1.

Advanced in vitro diagnosis technologies are highly desirable in early detection, prognosis, and progression monitoring of diseases. Here, we engineer a multiplex protein biosensing strategy based on the tunable liquid confinement self-assembly of multi-material heterochains, which show improved sensitivity, throughput, and accuracy compared to standard ELISA kits. By controlling the material combination and the number of ligand nanoparticles (NPs), we observe robust near-field enhancement as well as both strong electromagnetic resonance in polymer-semiconductor heterochains. In particular, their optical signals show a linear response to the coordination number of the semiconductor NPs in a wide range. Accordingly, a visible nanophotonic biosensor is developed by functionalizing antibodies on central polymer chains that can identify target proteins attached to semiconductor NPs. This allows for the specific detection of multiple protein biomarkers from healthy people and pancreatic cancer patients in one step with an ultralow detection limit (1 pg/mL). Furthermore, rapid and high-throughput quantification of protein expression levels in diverse clinical samples such as buffer, urine, and serum is achieved by combining a neural network algorithm, with an average accuracy of 97.3%. This work demonstrates that the heterochain-based biosensor is an exemplary candidate for constructing next-generation diagnostic tools and suitable for many clinical settings.

Microalgae metabolite analysis is fundamental for the rational design of metabolic engineering strategies for the biosynthesis of high-value products. Mass spectrometry (MS) has been utilized for single-cell microalgae analysis. However, limitations in the detection throughput and polarities of detectable substances make it difficult to realize high-throughput screening of high-performance microalgae. Herein, a plasma-assisted label-free mass cytometry, named as PACyESI-MS, was proposed combining the advantages of orthogonal hybrid ionization and high-throughput MS analysis, which realized rapid metabolite detection of single microalgae. The cell detection throughput of PACyESI-MS was up to 52 cells/min. Dozens of the critical primary and secondary metabolites within single microalgae were detected simultaneously, including pigments, lipids, and energy metabolites. Furthermore, metabolite changes of Chlamydomonas reinhardtii and Haematococcus pluvialis under nitrogen deficiency stress were studied. Discrimination of Chlamydomonas under different nutrient conditions was realized using single-cell metabolite profiles obtained by PACyESI-MS. The relationships between the accumulation of bioactive astaxanthin and changes in functional primary metabolites of Haematococcus were investigated. It was demonstrated that PACyESI-MS can detect the flexible change of metabolites in single microalgae cells under different nutritional conditions and during the synthesis of high-value products, which is expected to become an important tool for the design of metabolic engineering-based high-performance microalgae factories.

In this work, we reported a cholesterol oxidase (Chox)-loaded platinum (Pt) nanozyme with the collaborative cascade nanoreactor for the construction of nanozyme-enzyme-linked immunosorbent assay (N-ELSA) models to realize high-throughput rapid evaluation of cancer markers. Considering the high specific surface area and manipulable surface sites, ZIF-8 was used as a substrate for natural enzyme and nanozyme loading. The constructed ZIF-8-Pt nanozyme platform exhibited efficient enzyme-like catalytic efficiency with a standard corrected activity of 60.59 U mg -1 , which was 12 times higher than that of the ZIF-8 precursor, and highly efficient photothermal conversion efficiency (∼35.49%). In N-ELISA testing, developed multienzyme photothermal probes were immobilized in microplates based on antigen-antibody-specific reactions. Cholesterol was reacted in a cascade to reactive oxygen radicals, which attacked 3,3',5,5'-tetramethylbenzidine, causing it to oxidize and color change, thus exhibiting highly enhanced efficient photothermal properties. Systematic temperature evaluations were performed by a hand-held microelectromechanical system thermal imager under the excitation of an 808 nm surface light source to determine the cancer antigen 15-3 (CA15-3) profiles in the samples. Encouragingly, the temperature signal from the microwells increased with increasing CA15-3, with a linear range of 2 mU mL -1 to 100 U mL -1 , considering it to be the sensor with the widest working range for visualization and portability available. This work provides new horizons for the development of efficient multienzyme portable colorimetric-photothermal platforms to help advance the community-based process of early cancer detection.

Alzheimer's disease (AD) starts decades before cognitive symptoms develop. Easily accessible and cost-effective biomarkers that accurately reflect AD pathology are essential for both monitoring and therapeutics of AD. Neurofilament light chain (NfL) levels in blood and cerebrospinal fluid are increased in AD more than a decade before the expected onset, thus providing one of the most promising blood biomarkers for monitoring of AD. The clinical practice of employing single-molecule array (Simoa) technology for routine use in patient care is limited by the high costs. Herein, we developed a microarray chip-based high-throughput screening method and screened an attractive self-assembling peptide targeting NfL. Through directly "imprinting" and further analyzing the sequences, morphology, and affinity of the identified self-assembling peptides, the Pep-NfL peptide nanosheet with high binding affinity toward NfL ( K D = 1.39 × 10 -9 mol/L), high specificity, and low cost was characterized. The superior binding ability of Pep-NfL was confirmed in AD mouse models and cell lines. In the clinical setting, the Pep-NfL peptide nanosheets hold great potential for discriminating between patients with AD ( P < 0.001, n = 37), mild cognitive impairment ( P < 0.05, n = 26), and control groups ( n = 30). This work provides a high-throughput, high-sensitivity, and economical system for noninvasive tracking of AD to monitor neurodegeneration at different stages of disease. The obtained Pep-NfL peptide nanosheet may be useful for assessing dynamic changes in plasma NfL concentrations to evaluate disease-modifying therapies as a surrogate end point of neurodegeneration in clinical trials.

Efficient cell manipulation is essential for numerous applications in bioanalysis and medical diagnosis. However, the lack of stability and strength in the secondary flow, coupled with the narrow range of practical throughput, severely restricts the diverse applications. Herein, we present an innovative inertial microfluidic device that employs a spiral channel for high-throughput cell manipulation. Our investigation demonstrates that the regulation of Dean-like secondary flow in the microchannel can be achieved through geometric confinement. Introducing ordered microstructures into the ultralong spiral channel (>90 cm) stabilizes and accelerates the secondary flow among different loops. Consequently, effective manipulation of blood cells within a wide cell throughput range (1.73 × 10 8 to 1.16 × 10 9 cells/min) and cancer cells across a broad throughput range (0.5 × 10 6 to 5 × 10 7 cells/min) can be achieved. In comparison to previously reported technologies, our engineering approach of stabilizing and accelerating secondary flow offers specific performance for cell manipulation under a wide range of high-throughput manner. This engineered spiral channel would be promising in biomedical analysis, especially when cells need to be focused efficiently on large-volume liquid samples.

The ability to conduct effective high throughput screening (HTS) campaigns in drug discovery is often hampered by the detection of false positives in these assays due to small colloidally aggregating molecules (SCAMs). SCAMs can produce artifactual hits in HTS by nonspecific inhibition of the protein target. In this work, we present a new computational prediction tool for detecting SCAMs based on their 2D chemical structure. The tool, called the boosted aggregation detection (BAD) molecule filter, employs decision tree ensemble methods, namely, the CatBoost classifier and the light gradient-boosting machine, to significantly improve the detection of SCAMs. In developing the filter, we explore models trained on individual data sets, a consensus approach using these models, and, third, a merged data set approach, each tailored for specific drug discovery needs. The individual data set method emerged as most effective, achieving 93% sensitivity and 90% specificity, outperforming existing state-of-the-art models by 20 and 5%, respectively. The consensus models offer broader chemical space coverage, exceeding 90% for all testing sets. This feature is an important aspect particularly for early stage medicinal chemistry projects, and provides information on applicability domain. Meanwhile, the merged data set models demonstrated robust performance, with a notable sensitivity of 79% in the comprehensive 10-fold cross-validation test set. A SHAP analysis of model features indicates the importance of hydrophobicity and molecular complexity as primary factors influencing the aggregation propensity. The BAD molecule filter is readily accessible for the public usage on https://molmodlab-aau.com/Tools.html. This filter provides a new, more robust tool for aggregate prediction in the early stages of drug discovery to optimize hit rates and reduce associated testing and validation overheads.

Immune modulation through the intracellular delivery of nucleoside-modified mRNA to immune cells is an attractive approach for in vivo immunoengineering, with applications in infectious disease, cancer immunotherapy, and beyond. Lipid nanoparticles (LNPs) have come to the fore as a promising nucleic acid delivery platform, but LNP design criteria remain poorly defined, making the rate-limiting step for LNP discovery the screening process. In this study, we employed high-throughput in vivo LNP screening based on molecular barcoding to investigate the influence of LNP composition on immune tropism with applications in vaccines and systemic immunotherapies. Screening a large LNP library under both intramuscular ( i.m. ) and intravenous ( i.v. ) injection, we observed differential influences on LNP uptake by immune populations across the two administration routes, gleaning insight into LNP design criteria for in vivo immunoengineering. In validation studies, the lead LNP formulation for i.m. administration demonstrated substantial mRNA translation in the spleen and draining lymph nodes with a more favorable biodistribution profile than LNPs formulated with the clinical standard ionizable lipid DLin-MC3-DMA (MC3). The lead LNP formulations for i.v. administration displayed potent immune transfection in the spleen and peripheral blood, with one lead LNP demonstrating substantial transfection of splenic dendritic cells and another inducing substantial transfection of circulating monocytes. Altogether, the immunotropic LNPs identified by high-throughput in vivo screening demonstrated significant promise for both locally- and systemically-delivered mRNA and confirmed the value of the LNP design criteria gleaned from our screening process, which could potentially inform future endeavors in mRNA vaccine and immunotherapy applications.

Hypoxia is a representative tumor characteristic associated with malignant progression in clinical patients. Engineered in vitro models have led to significant advances in cancer research, allowing for the investigation of cells in physiological environments and the study of disease mechanisms and processes with enhanced relevance. In this study, we propose a U-shape pillar strip for a 3D cell-lumped organoid model (3D-COM) to study the effects of hypoxia on lung cancer in a high-throughput manner. We developed a U-pillar strip that facilitates the aggregation of PDCs mixed with an extracellular matrix to make the 3D-COM in 384-plate array form. The response to three hypoxia-activated prodrugs was higher in the 3D-COM than in the 2D culture model. The protein expression of hypoxia-inducible factor 1 alpha (HIF-1α) and HIF-2α, which are markers of hypoxia, was also higher in the 3D-COM than in the 2D culture. The results show that 3D-COM better recapitulated the hypoxic conditions of lung cancer tumors than the 2D culture. Therefore, the U-shape pillar strip for 3D-COM is a good tool to study the effects of hypoxia on lung cancer in a high-throughput manner, which can efficiently develop new drugs targeting hypoxic tumors.

The precise prediction of molecular properties can greatly accelerate the development of new drugs. However, in silico molecular property prediction approaches have been limited so far to assays for which large amounts of data are available. In this study, we develop a new computational approach leveraging both the textual description of the assay of interest and the chemical structure of target compounds. By combining these two sources of information via self-supervised learning, our tool can provide accurate predictions for assays where no measurements are available. Remarkably, our approach achieves state-of-the-art performance on the FS-Mol benchmark for zero-shot prediction, outperforming a wide variety of deep learning approaches. Additionally, we demonstrate how our tool can be used for tailoring screening libraries for the assay of interest, showing promising performance in a retrospective case study on a high-throughput screening campaign. By accelerating the early identification of active molecules in drug discovery and development, this method has the potential to streamline the identification of novel therapeutics.

This Technical Note describes a dual-column liquid chromatography system coupled to mass spectrometry (LC-MS) for high-throughput bottom-up proteomic analysis. This system made full use of two 2-position 10-port valves and a binary pump with an integrated loading pump of a commercial LC instrument to provide successive operation of two parallel subsystems. Each subsystem consisted of a set of trap columns and an analytical column. A T-junction union was used to split the mobile phase from the loading pump into two parts. This allowed one set of columns to be washed and equilibrated, followed by the injection of the next sample, while the previous sample was eluting and being analyzed on the other set of columns, thereby greatly increasing the analysis throughput. This approach showed high reproducibility for the analysis of HeLa tryptic digests with average relative standard deviation (RSD) values of 1.75%, 6.90%, and 5.19% for the identification number of proteins, peptides, and peptide-spectrum matches (PSMs), respectively, across 10 consecutive runs. The capacity for peptide and protein identification, as well as proteome depth, of the dual-column LC system was comparable to a conventional single-column system. Due to its simple equipment requirements and set up process, this method should be highly accessible for other laboratories.

The worldwide spread of the metallo-β-lactamases (MBL), especially New Delhi metallo-β-lactamase-1 (NDM-1), is threatening the efficacy of β-lactams, which are the most potent and prescribed class of antibiotics in the clinic. Currently, FDA-approved MBL inhibitors are lacking in the clinic even though many strategies have been used in inhibitor development, including quantitative high-throughput screening (qHTS), fragment-based drug discovery (FBDD), and molecular docking. Herein, a machine learning-based prediction tool is described, which was generated using results from HTS of a large chemical library and previously published inhibition data. The prediction tool was then used for virtual screening of the NIH Genesis library, which was subsequently screened using qHTS. A novel MBL inhibitor was identified and shown to lower minimum inhibitory concentrations (MICs) of Meropenem for a panel of E. coli and K. pneumoniae clinical isolates expressing NDM-1. The mechanism of inhibition of this novel scaffold was probed utilizing equilibrium dialyses with metal analyses, native state electrospray ionization mass spectrometry, UV-vis spectrophotometry, and molecular docking. The uncovered inhibitor, compound 72922413, was shown to be 9-hydroxy-3-[(5-hydroxy-1-oxa-9-azaspiro[5.5]undec-9-yl)carbonyl]-4 H -pyrido[1,2- a ]pyrimidin-4-one.

Raman spectroscopy can provide nonbiased single-cell analysis based on the endogenous ensemble of biomolecules, with alterations in cellular content indicative of cell state and disease. The measurements themselves can be performed in a variety of modes: generally, full imaging takes the most time but can provide the most information. By reducing the imaging resolution and generating the most characteristic single-cell Raman spectrum in the shortest time, we optimize the utility of the Raman measurement for cell phenotyping. Here, we establish methods to compare these different measurement approaches and assess what, if any, undesired effects occur in the cell. Assuming that laser-induced damage should be apparent as a change in molecular spectra across sequential measurements, and by defining the information content as the Raman-based separability of two cell lines, we thereby establish a parameter range for optimum measurement sensitivity and single-cell throughput in single-cell Raman spectroscopic analysis. While the work here uses 532 nm irradiation, the same approach can be generalized to Raman analysis at other wavelengths.

Precise early diagnosis and staging are conducive to improving the prognosis of colorectal cancer (CRC) and gastric cancer (GC) patients. However, due to intrusive inspections and limited sensitivity, the prevailing diagnostic methods impede precisely large-scale screening. In this work, we reported a high-throughput serum metabolic patterns (SMP) screening strategy based on covalent organic frameworks-assisted laser desorption/ionization mass spectrometry (hf-COFsLDI-MS) for early diagnosis and staging of CRC and GC. Notably, 473 high-quality SMP were extracted without any tedious sample pretreatment and coupled with multiple machine learning algorithms; the area under the curve (AUC) value is 0.938 with 96.9% sensitivity for early CRC diagnosis, and the AUC value is 0.974 with 100% sensitivity for early GC diagnosis. Besides, the discrimination of CRC and GC is accomplished with an AUC value of 0.966 for the validation set. Also, the screened-out features were identified by MS/MS experiments, and 8 metabolites were identified as the biomarkers for CRC and GC. Finally, the corresponding disordered metabolic pathways were revealed, and the staging of CRC and GC was completed. This work provides an alternative high-throughput screening strategy for CRC and GC and highlights the potential of metabolic molecular diagnosis in clinical applications.

The advent of ultra-large libraries of drug-like compounds has significantly broadened the possibilities in structure-based virtual screening, accelerating the discovery and optimization of high-quality lead chemotypes for diverse clinical targets. Compared to traditional high-throughput screening, which is constrained to libraries of approximately one million compounds, the ultra-large virtual screening approach offers substantial advantages in both cost and time efficiency. By expanding the chemical space with compounds synthesized from easily accessible and reproducible reactions and utilizing a large, diverse set of building blocks, we can enhance both the diversity and quality of the discovered lead chemotypes. In this study, we explore new chemical spaces using reactions of sulfur(VI) fluorides to create a combinatorial library consisting of several hundred million compounds. We screened this virtual library for cannabinoid type II receptor (CB 2 ) antagonists using the high-resolution structure in conjunction with a rationally designed antagonist, AM10257. The top-predicted compounds were then synthesized and tested in vitro for CB 2 binding and functional antagonism, achieving an experimentally validated hit rate of 55%. Our findings demonstrate the effectiveness of reliable reactions, such as sulfur fluoride exchange, in diversifying ultra-large chemical spaces and facilitate the discovery of new lead compounds for important biological targets.

The global management for persistent, mobile, and toxic (PMT) and very persistent and very mobile (vPvM) substances has been further strengthened with the rapid increase of emerging contaminants. The development of a ready-to-use and publicly available tool for the high-throughput screening of PMT/vPvM substances is thus urgently needed. However, the current model building with the coupling of conventional algorithms, small-scale data set, and simplistic features hinders the development of a robust model for screening PMT/vPvM with wide application domains. Here, we construct a graph convolutional network (GCN)-enhanced model with feature fusion of a molecular graph and molecular descriptors to effectively utilize the significant correlation between critical descriptors and PMT/vPvM substances. The model is built with 213,084 substances following the latest PMT classification criteria. The application domains of the GCN-enhanced model assessed by kernel density estimation demonstrate the high suitability for high-throughput screening PMT/vPvM substances with both a high accuracy rate (86.6%) and a low false-negative rate (6.8%). An online server named PMT/vPvM profiler is further developed with a user-friendly web interface (http://www.pmt.zj.cn/). Our study facilitates a more efficient evaluation of PMT/vPvM substances with a globally accessible screening platform.

Recently emerging generative AI models enable us to produce a vast number of compounds for potential applications. While they can provide novel molecular structures, the synthetic feasibility of the generated molecules is often questioned. To address this issue, a few recent studies have attempted to use deep learning models to estimate the synthetic accessibility of many molecules rapidly. However, retrosynthetic analysis tools used to train the models rely on reaction templates automatically extracted from a large reaction database that are not domain-specific and may exhibit low chemical correctness. To overcome this limitation, we introduce DFRscore (Drug-Focused Retrosynthetic score), a deep learning-based approach for a more practical assessment of synthetic accessibility in drug discovery. The DFRscore model is trained exclusively on drug-focused reactions, providing a predicted number of minimally required synthetic steps for each compound. This approach enables practitioners to filter out compounds that do not meet their desired level of synthetic accessibility at an early stage of high-throughput virtual screening for accelerated drug discovery. The proposed strategy can be easily adapted to other domains by adjusting the synthesis planning setup of the reaction templates and starting materials.

Recent advances have demonstrated the promise of complex multicomponent polymeric supports to enable supra-biological enzyme performance. However, the discovery of such supports has been limited by time-consuming, low-throughput synthesis and screening. Here, we describe a novel combinatorial and high-throughput platform that enables rapid screening of complex and heterogeneous copolymer brushes as enzyme immobilization supports, named combinatorial high-throughput enzyme support screening (CHESS). Using a 384-well plate format, we synthesized arrays of three-component polymer brushes in the microwells using photoactivated surface-initiated polymerization and immobilized enzymes in situ. The utility of CHESS to identify optimal immobilization supports under thermally and chemically denaturing conditions was demonstrated using Bacillus subtilis Lipase A (LipA). The identification of supports with optimal compositions was validated by immobilizing LipA on polymer-brush-modified biocatalyst particles. We further demonstrated that CHESS could be used to predict the optimal composition of polymer brushes a priori for the previously unexplored enzyme, alkaline phosphatase (AlkP). Our findings demonstrate that CHESS represents a predictable and reliable platform for dramatically accelerating the search of chemical compositions for immobilization supports and further facilitates the discovery of biocompatible and stabilizing materials.

G protein-coupled receptors (GPCRs) are central mediators of cell signaling and physiological function. Despite their biological significance, GPCRs have not been widely studied in the field of toxicology. Herein, we investigated these receptors as novel targets of plastic chemicals using a high-throughput drug screening assay with 126 human non-olfactory GPCRs. In a first-pass screen, we tested the activity of triphenol phosphate, bisphenol A, and diethyl phthalate, as well as three real-world mixtures of chemicals extracted from plastic food packaging covering all major polymer types. We found 11 GPCR-chemical interactions, of which the chemical mixtures exhibited the most robust activity at adenosine receptor 1 (ADORA1) and melatonin receptor 1 (MTNR1A). We further confirm that polyvinyl chloride and polyurethane products contain ADORA1 or MTNRA1 agonists using a confirmatory secondary screen and pharmacological knockdown experiments. Finally, an analysis of the associated gene ontology terms suggests that ADORA1 and MTNR1A activation may be linked to downstream effects on circadian and metabolic processes. This work highlights that signaling disruption caused by plastic chemicals is broader than that previously believed and demonstrates the relevance of nongenomic pathways, which have, thus far, remained unexplored.

Chitinase-3-like-1 (CHI3L1), also known as YKL-40, is a glycoprotein linked to inflammation, fibrosis, and cancer. This study explored CHI3L1's interactions with various oligosaccharides using microscale thermophoresis (MST) and AlphaScreen (AS). These investigations guided the development of high-throughput screening assays to assess interference of small molecules in binding between CHI3L1 and biotinylated small molecules or heparan sulfate-based probes. Small molecule binders of YKL-40 were identified in our chitotriosidase inhibitors library with MST and confirmed through X-ray crystallography. Based on cocrystal structures of potent hit compounds with CHI3L1, small molecule probes 19 and 20 were designed for an AS assay. Structure-based optimization led to compounds 30 and 31 with nanomolar activities and drug-like properties. Additionally, an orthogonal AS assay using biotinylated heparan sulfate as a probe was developed. The compounds' affinity showed a significant correlation in both assays. These screening tools and compounds offer novel avenues for investigating the role of CHI3L1.

Drug development is a complex, costly, and time-consuming endeavor. While high-throughput screening (HTS) plays a critical role in the discovery stage, it is one of many factors contributing to these challenges. In certain contexts, virtual screening can complement the HTS, potentially offering a more streamlined approach in the initial stages of drug discovery. Molecular docking is an example of a popular virtual screening technique that is often used for this purpose; however, its effectiveness can vary greatly. This has led to the use of consensus docking approaches that combine results from different docking methods to improve the identification of active compounds and reduce the occurrence of false positives. However, many of these methods do not fully leverage the latest advancements in molecular docking. In response, we present ESSENCE-Dock (Effective Structural Screening ENrichment ConsEnsus Dock), a new consensus docking workflow aimed at decreasing false positives and increasing the discovery of active compounds. By utilizing a combination of novel docking algorithms, we improve the selection process for potential active compounds. ESSENCE-Dock has been made to be user-friendly, requiring only a few simple commands to perform a complete screening while also being designed for use in high-performance computing (HPC) environments.

A simple approach was developed to computationally construct a polymer dataset by combining simplified molecular-input line-entry system (SMILES) strings of a targeted polymer backbone and a variety of molecular fragments. This method was used to create 14 polymer datasets by combining seven polymer backbones and molecules from two large molecular datasets (MOSES and QM9). Polymer backbones that were studied include four polydimethylsiloxane (PDMS) based backbones, poly(ethylene oxide) (PEO), poly(allyl glycidyl ether) (PAGE), and polyphosphazene (PPZ). The generated polymer datasets can be used for various cheminformatics tasks, including high-throughput screening for gas permeability and selectivity. This study utilized machine learning (ML) models to screen the polymers for CO 2 /CH 4 and CO 2 /N 2 gas separation using membranes. Several polymers of interest were identified. The results highlight that employing an ML model fitted to polymer selectivities leads to higher accuracy in predicting polymer selectivity compared to using the ratio of predicted permeabilities.

We herein describe the development and application of a modular technology platform which incorporates recent advances in plate-based microscale chemistry, automated purification, in situ quantification, and robotic liquid handling to enable rapid access to high-quality chemical matter already formatted for assays. In using microscale chemistry and thus consuming minimal chemical matter, the platform is not only efficient but also follows green chemistry principles. By reorienting existing high-throughput assay technology, the platform can generate a full package of relevant data on each set of compounds in every learning cycle. The multiparameter exploration of chemical and property space is hereby driven by active learning models. The enhanced compound optimization process is generating knowledge for drug discovery projects in a time frame never before possible.

Interleukin-1 receptor-associated kinase 4 (IRAK4) plays a critical role in innate inflammatory processes. Here, we describe the discovery of two clinical candidate IRAK4 inhibitors, BAY1834845 (zabedosertib) and BAY1830839 , starting from a high-throughput screening hit derived from Bayer's compound library. By exploiting binding site features distinct to IRAK4 using an in-house docking model, liabilities of the original hit could surprisingly be overcome to confer both candidates with a unique combination of good potency and selectivity. Favorable DMPK profiles and activity in animal inflammation models led to the selection of these two compounds for clinical development in patients.

Droplet-based high-throughput screening systems are an emerging technology that provides a quick test to screen millions of cells with distinctive characteristics. Biopharmaceuticals, specifically therapeutic proteins, are produced by culturing cells that secrete heterologous recombinant proteins with different populations and expression levels; therefore, a technology to discriminate cells that produce more target proteins is needed. Here, we present a droplet-based microfluidic strategy for encapsulating, screening, and selecting target cells with redox-responsive hydrogel beads (HBs). As a proof-of-concept study, we demonstrate the enrichment of hybridoma cells with enhanced capability of antibody secretion using horseradish peroxidase (HRP)-catalyzed hydrogelation of tetra-thiolate poly(ethylene glycol); hybridoma cells were encapsulated in disulfide-bonded HBs. Recombinant protein G or protein M with a C-terminal cysteine residue was installed in the HBs via disulfide bonding to capture antibodies secreted from the cells. HBs were fluorescently stained by adding the protein L-HRP conjugate using a tyramide signal amplification system. HBs were then separated by fluorescence-activated droplet sorting and degraded by reducing the disulfide bonds to recover the target cells. Finally, we succeeded in the selection of hybridoma cells with enhanced antibody secretion, indicating the potential of this system in the therapeutic protein production.

Quorum sensing (QS) is a cell-cell communication mechanism by which bacteria synchronize social behaviors such as biofilm formation and virulence factor secretion by producing and sensing small molecular signals. Quorum quenching (QQ) by degrading signals or blocking signal transmissions has become a promising strategy for disrupting QS and preventing bacterial infection and biofilm formation. However, studies of high-throughput screening and identification approaches for quorum-sensing inhibitors (QSIs) are still inadequate. In this work, we developed a sensitive, high-throughput approach for screening QSIs based on the bacterial biosensor strain Agrobacterium tumefaciens N5 (pBA7P), which contains a traG gene promoter induced by QS signals fused with a promoterless β-lactamase gene reporter. Using this approach, we identified 31 QQ bacteria from ∼2000 soil bacterial isolates, some belonging to the genera Bosea , Cupriavidus , and Flavobacterium that have not been reported previously as QQ bacteria. We also identified four QS inhibitory compounds and one QS signal analogue from ∼5000 small-molecule compounds, which profoundly affected the expression of QS-regulated genes and phenotypes of the pathogenic bacteria. This high-throughput screening system is effective and sensitive for screening of both QQ microbes and small molecules, enabling the discovery of a wide variety of biocompatible compounds.

Advanced computational methods are being actively sought to address the challenges associated with the discovery and development of new combinatorial materials, such as formulations. A widely adopted approach involves domain-informed high-throughput screening of individual components that can be combined together to form a formulation. This manages to accelerate the discovery of new compounds for a target application but still leaves the process of identifying the right "formulation" from the shortlisted chemical space largely a laboratory experiment-driven process. We report a deep learning model, the Formulation Graph Convolution Network (F-GCN), that can map the structure-composition relationship of the formulation constituents to the property of liquid formulation as a whole. Multiple GCNs are assembled in parallel that featurize formulation constituents domain-intuitively on the fly. The resulting molecular descriptors are scaled based on the respective constituent's molar percentage in the formulation, followed by integration into a combined formulation descriptor that represents the complete formulation to an external learning architecture. The use case of the proposed formulation learning model is demonstrated for battery electrolytes by training and testing it on two exemplary data sets representing electrolyte formulations vs battery performance: one data set is sourced from the literature about Li/Cu half-cells, while the other is obtained by lab experiments related to lithium-iodide full-cell chemistry. The model is shown to predict performance metrics such as Coulombic efficiency (CE) and specific capacity of new electrolyte formulations with the lowest reported errors. The best-performing F-GCN model uses molecular descriptors derived from molecular graphs (GCNs) that are informed with HOMO-LUMO and electric moment properties of the molecules using a knowledge transfer technique.

Broad-spectrum anti-infective chemotherapy agents with activity against Trypanosomes , Leishmania, and Mycobacterium tuberculosis species were identified from a high-throughput phenotypic screening program of the 456 compounds belonging to the Ty-Box, an in-house industry database. Compound characterization using machine learning approaches enabled the identification and synthesis of 44 compounds with broad-spectrum antiparasitic activity and minimal toxicity against Trypanosoma brucei , Leishmania Infantum, and Trypanosoma cruzi . In vitro studies confirmed the predictive models identified in compound 40 which emerged as a new lead, featured by an innovative N -(5-pyrimidinyl)benzenesulfonamide scaffold and promising low micromolar activity against two parasites and low toxicity. Given the volume and complexity of data generated by the diverse high-throughput screening assays performed on the compounds of the Ty-Box library, the chemoinformatic and machine learning tools enabled the selection of compounds eligible for further evaluation of their biological and toxicological activities and aided in the decision-making process toward the design and optimization of the identified lead.

Tricyclic tetrahydroquinolines (THQs) have been repeatedly reported as hits across a diverse range of high-throughput screening (HTS) campaigns. The activities of these compounds, however, are likely due to reactive byproducts that interfere with the assay. As a lesser studied class of pan-assay interference compounds, the mechanism by which fused THQs react with protein targets remains largely unknown. During HTS follow-up, we characterized the behavior and stability of several fused tricyclic THQs. We synthesized key analogues to pinpoint the cyclopentene ring double bond as a source of reactivity of fused THQs. We found that these compounds degrade in solution under standard laboratory conditions in days. Importantly, these observations make it likely that fused THQs, which are ubiquitously found within small molecule screening libraries, are unlikely the intact parent compounds. We urge deprioritization of tricylic THQ hits in HTS follow-up and caution against the investment of resources to follow-up on these problematic compounds.

Colloidal stability is an important consideration when developing high concentration mAb formulations. PEG-induced protein precipitation is a commonly used assay to assess the colloidal stability of protein solutions. However, the practical usefulness and the current theoretical model for this assay have yet to be verified over a large formulation space across multiple mAbs and mAb-based modalities. In the present study, we used PEG-induced protein precipitation assays to evaluate colloidal stability of 3 mAbs in 24 common formulation buffers at 20 and 5 °C. These prediction assays were conducted at low protein concentration (1 mg/mL). We also directly characterized high concentration (100 mg/mL) formulations for cold-induced phase separation, turbidity, and concentratibility by ultrafiltration. This systematic study allowed analysis of the correlation between the results of low concentration assays and the high concentration attributes. The key findings of this study include the following: (1) verification of the usefulness of three different parameters ( C mid , μ B , and T cloud ) from PEG-induced protein precipitation assays for ranking colloidal stability of high concentration mAb formulations; (2) a new method to implement PEG-induced protein precipitation assay suitable for high throughput screening with low sample consumption; (3) improvement in the theoretical model for calculating robust thermodynamic parameters of colloidal stability (μ B and ε B ) that are independent of specific experimental settings; (4) systematic evaluation of the effects of pH and buffer salts on colloidal stability of mAbs in common formulation buffers. These findings provide improved theoretical and practical tools for assessing the colloidal stability of mAbs and mAb-based modalities during formulation development.

Human carboxylesterase 2A (hCES2A) is an important endoplasmic reticulum (ER)-resident enzyme that is responsible for the hydrolytic metabolism or activation of numerous ester-bearing drugs and environmental toxins. The previously reported hCES2A fluorogenic substrates suffer from limited emission wavelength, low specificity, and poor localization accuracy, thereby greatly limiting the in situ functional imaging of hCES2A and drug discovery. Herein, a rational ligand design strategy was adopted to construct a highly specific near-infrared (NIR) substrate for hCES2A. Following scaffold screening and recognition group optimization, HTCF was identified as a desirable NIR fluorophore with excellent photophysical properties and high ER accumulation ability, while several HTCF esters held a high potential to be good hCES2A substrates. Further investigations revealed that TP-HTCF (the tert -pentyl ester of HTCF ) was an ideal substrate with ultrahigh sensitivity, excellent specificity, and a substantial signal-to-noise ratio. Upon the addition of hCES2A, TP-HTCF could be rapidly hydrolyzed to release HTCF , a chemically stable product that emitted bright fluorescent signals at around 670 nm. A TP-HTCF -based biochemical assay was then established for the high-throughput screening of potent and cell-active hCES2A inhibitors from an in-house compound library. Furthermore, TP-HTCF displayed high imaging resolution for imaging hCES2A in living cells as well as mouse liver slices and tumor-xenograft mice. Collectively, this study demonstrates a rational strategy for developing highly specific fluorogenic substrates for an ER-resident target enzyme, while TP-HTCF can act as a practical tool for sensing hCES2A in living systems.

Glass is by far the most common substrate for biomolecular arrays, including high-throughput sequencing flow cells and microarrays. The native glass hydroxyl surface is modified by using silane chemistry to provide appropriate functional groups and reactivities for either in situ synthesis or surface immobilization of biologically or chemically synthesized biomolecules. These arrays, typically of oligonucleotides or peptides, are then subjected to long incubation times in warm aqueous buffers prior to fluorescence readout. Under these conditions, the siloxy bonds to the glass are susceptible to hydrolysis, resulting in significant loss of biomolecules and concomitant loss of signal from the assay. Here, we demonstrate that functionalization of glass surfaces with dipodal silanes results in greatly improved stability compared to equivalent functionalization with standard monopodal silanes. Using photolithographic in situ synthesis of DNA, we show that dipodal silanes are compatible with phosphoramidite chemistry and that hybridization performed on the resulting arrays provides greatly improved signal and signal-to-noise ratios compared with surfaces functionalized with monopodal silanes.

A method is introduced for the automated analysis of reactivity exploration for extended in silico databases of transition-metal catalysts. The proposed workflow is designed to tackle two key challenges for bias-free mechanistic explorations on large databases of catalysts: (1) automated exploration of the chemical space around each catalyst with unique structural and chemical features and (2) automated analysis of the resulting large chemical data sets. To address these challenges, we have extended the application of our previously developed ReNeGate method for bias-free reactivity exploration and implemented an automated analysis procedure to identify the classes of reactivity patterns within specific catalyst groups. Our procedure applied to an extended series of representative Mn(I) pincer complexes revealed correlations between structural and reactive features, pointing to new channels for catalyst transformation under the reaction conditions. Such an automated high-throughput virtual screening of systematically generated hypothetical catalyst data sets opens new opportunities for the design of high-performance catalysts as well as an accelerated method for expert bias-free high-throughput in silico reactivity exploration.

The emergence of ultra-large screening libraries, filled to the brim with billions of readily available compounds, poses a growing challenge for docking-based virtual screening. Machine learning (ML)-boosted strategies like the tool HASTEN combine rapid ML prediction with the brute-force docking of small fractions of such libraries to increase screening throughput and take on giga-scale libraries. In our case study of an anti-bacterial chaperone and an anti-viral kinase, we first generated a brute-force docking baseline for 1.56 billion compounds in the Enamine REAL lead-like library with the fast Glide high-throughput virtual screening protocol. With HASTEN, we observed robust recall of 90% of the true 1000 top-scoring virtual hits in both targets when docking only 1% of the entire library. This reduction of the required docking experiments by 99% significantly shortens the screening time. In the kinase target, the employment of a hydrogen bonding constraint resulted in a major proportion of unsuccessful docking attempts and hampered ML predictions. We demonstrate the optimization potential in the treatment of failed compounds when performing ML-boosted screening and benchmark and showcase HASTEN as a fast and robust tool in a growing arsenal of approaches to unlock the chemical space covered by giga-scale screening libraries for everyday drug discovery campaigns.

An important goal for bottom-up synthetic biology is to construct tissue-like structures from artificial cells. The key is the ability to control the assembly of the individual artificial cells. Unlike most methods resorting to external fields or sophisticated devices, inspired by the hanging drop method used for culturing spheroids of biological cells, we employ a capillary-driven approach to assemble giant unilamellar vesicles (GUVs)-based protocells into colonized prototissue arrays by means of a coverslip with patterned wettability. By spatially confining and controllably merging a mixed population of lipid-coated double-emulsion droplets that hang on a water/oil interface, an array of synthetic tissue-like constructs can be obtained. Each prototissue module in the array comprises multiple tightly packed droplet compartments where interfacial lipid bilayers are self-assembled at the interfaces both between two neighboring droplets and between the droplet and the external aqueous environment. The number, shape, and composition of the interconnected droplet compartments can be precisely controlled. Each prototissue module functions as a processer, in which fast signal transports of molecules via cell-cell and cell-environment communications have been demonstrated by molecular diffusions and cascade enzyme reactions, exhibiting the ability to be used as biochemical sensing and microreactor arrays. Our work provides a simple yet scalable and programmable method to form arrays of prototissues for synthetic biology, tissue engineering, and high-throughput assays.

Assessing the prospective climate preservation potential of novel, innovative, but immature chemical production techniques is limited by the high number of process synthesis options and the lack of reliable, high-throughput quantitative sustainability pre-screening methods. This study presents the sequential use of data-driven hybrid prediction (ANN-RSM-DOM) to streamline waste-to-dimethyl ether (DME) upcycling using a set of sustainability criteria. Artificial neural networks (ANNs) are developed to generate in silico waste valorization experimental results and ex-ante model the operating space of biorefineries applying the organic fraction of municipal solid waste (OFMSW) and sewage sludge (SS). Aspen Plus process flowsheeting and ANN simulations are postprocessed using the response surface methodology (RSM) and desirability optimization method (DOM) to improve the in-depth mechanistic understanding of environmental systems and identify the most benign configurations. The hybrid prediction highlights the importance of targeted waste selection based on elemental composition and the need to design waste-specific DME synthesis to improve techno-economic and environmental performances. The developed framework reveals plant configurations with concurrent climate benefits (-1.241 and -2.128 kg CO 2 -eq (kg DME) -1 ) and low DME production costs (0.382 and 0.492 € (kg DME) -1 ) using OFMSW and SS feedstocks. Overall, the multi-scale explorative hybrid prediction facilitates early stage process synthesis, assists in the design of block units with nonlinear characteristics, resolves the simultaneous analysis of qualitative and quantitative variables, and enables the high-throughput sustainability screening of low technological readiness level processes.

Predicting adsorption energies of small molecules (e.g., OH, OOH, CO) on electrocatalysts involved in electrochemical reactions aids in accelerating the design and screening of electrocatalysts. Avoiding computationally expensive electronic structure calculations increases the speed of such predictions. Geometric and electronic descriptors have been reported to characterize the environment around surface active sites and predict adsorption energies. However, these descriptors cannot be used to predict adsorption energies of small molecules on various substrates, e.g., metal-oxide and nonmetal electrocatalysts. We compare the performance of these descriptors in predicting adsorption energies of small molecules on various electrocatalysts with adsorption energies calculated from density functional theory. We show that two recently developed machine learning algorithms, Crystal Graph Convolutional Neural Network (CGCNN) and Atomistic Line Graph Neural Network (ALIGNN), outperform the reported descriptors based on geometric (coordination number of the active site and its nearest neighbors) and electronic (the bond-energy-integrated orbitalwise coordination number, the electronegativity, and the number of valence electrons of the active site) properties in predicting the adsorption energies. Our results suggest that ALIGNN is almost always more accurate than CGCNN in adsorption energy predictions. The improvement ranges from 0.02 to 1.0 eV in the mean absolute errors (MAEs). We also compare the performance of CGCNN and ALIGNN algorithms in predicting the overpotentials of the oxygen evolution reaction occurring on various electrocatalysts with MAEs of 0.06 and 0.05 V, respectively.