An artificial intelligence platform for automated PFAS subgroup classification: A discovery tool for PFAS screening.

Since structural analyses and toxicity assessments have not been able to keep up with the discovery of unknown per- and polyfluoroalkyl substances (PFAS), there is an urgent need for effective categorization and grouping of PFAS. In this study, we presented PFAS-Atlas, an artificial intelligence-based platform containing a rule-based automatic classification system and a machine learning-based grouping model. Compared with previously developed classification software, the platform's classification system follows the latest Organization for Economic Co-operation and Development (OECD) definition of PFAS and reduces the number of uncategorized PFAS. In addition, the platform incorporates deep unsupervised learning models to visualize the chemical space of PFAS by clustering similar structures and linking related classes. Through real-world use cases, we demonstrate that PFAS-Atlas can rapidly screen for relationships between chemical structure and persistence, bioaccumulation, or toxicity data for PFAS. The platform can also guide the planning of the PFAS testing strategy by showing which PFAS classes urgently require further attention. Ultimately, the release of PFAS-Atlas will benefit both the PFAS research and regulation communities.

Connecting Vibrational Spectroscopy to Atomic Structure via Supervised Manifold Learning: Beyond Peak Analysis.

Vibrational spectroscopy is a nondestructive technique commonly used in chemical and physical analyses to determine atomic structures and associated properties. However, the evaluation and interpretation of spectroscopic profiles based on human-identifiable peaks can be difficult and convoluted. To address this challenge, we present a reliable protocol based on supervised manifold learning techniques meant to connect vibrational spectra to a variety of complex and diverse atomic structure configurations. As an illustration, we examined a large database of virtual vibrational spectroscopy profiles generated from atomistic simulations for silicon structures subjected to different stress, amorphization, and disordering states. We evaluated representative features in those spectra via various linear and nonlinear dimensionality reduction techniques and used the reduced representation of those features with decision trees to correlate them with structural information unavailable through classical human-identifiable peak analysis. We show that our trained model accurately (over 97% accuracy) and robustly (insensitive to noise) disentangles the contribution from the different material states, hence demonstrating a comprehensive decoding of spectroscopic profiles beyond classical (human-identifiable) peak analysis.

Data analytics-guided rational design of antimicrobial nanomedicines against opportunistic, resistant pathogens.

Nanoparticle carriers can improve antibiotic efficacy by altering drug biodistribution. However, traditional screening is impracticable due to a massive dataspace. A hybrid informatics approach was developed to identify polymer, antibiotic, and particle determinants of antimicrobial nanomedicine activity against Burkholderia cepacia, and to model nanomedicine performance. Polymer glass transition temperature, drug octanol-water partition coefficient, strongest acid dissociation constant, physiological charge, particle diameter, count and mass mean polydispersity index, zeta potential, fraction drug released at 2 h, and fraction release slope at 2 h were highly correlated with antimicrobial performance. Graph analysis provided dimensionality reduction while preserving nonlinear descriptor-property relationships, enabling accurate modeling of nanomedicine performance. The model successfully predicted particle performance in holdout validation, with moderate accuracy at rank-ordering. This data analytics-guided approach provides an important step toward the development of a rational design framework for antimicrobial nanomedicines against resistant infections by selecting appropriate carriers and payloads for improved potency.

RNA Nanovaccine Protects against White Spot Syndrome Virus in Shrimp.

In the last 15 years, crustacean fisheries have experienced billions of dollars in economic losses, primarily due to viral diseases caused by such pathogens as white spot syndrome virus (WSSV) in the Pacific white shrimp Litopenaeus vannamei and Asian tiger shrimp Penaeus monodon. To date, no effective measures are available to prevent or control disease outbreaks in these animals, despite their economic importance. Recently, double-stranded RNA-based vaccines have been shown to provide specific and robust protection against WSSV infection in cultured shrimp. However, the limited stability of double-stranded RNA is the most significant hurdle for the field application of these vaccines with respect to delivery within an aquatic system. Polyanhydride nanoparticles have been successfully used for the encapsulation and release of vaccine antigens. We have developed a double-stranded RNA-based nanovaccine for use in shrimp disease control with emphasis on the Pacific white shrimp L. vannamei. Nanoparticles based on copolymers of sebacic acid, 1,6-bis(p-carboxyphenoxy)hexane, and 1,8-bis(p-carboxyphenoxy)-3,6-dioxaoctane exhibited excellent safety profiles, as measured by shrimp survival and histological evaluation. Furthermore, the nanoparticles localized to tissue target replication sites for WSSV and persisted through 28 days postadministration. Finally, the nanovaccine provided ~80% protection in a lethal WSSV challenge model. This study demonstrates the exciting potential of a safe, effective, and field-applicable RNA nanovaccine that can be rationally designed against infectious diseases affecting aquaculture.

Arterial and venous thrombotic stroke after ChAdOx1 nCoV-19 vaccine.

Vaccine-induced thrombosis with thrombocytopenia (VITT) is a recently-described condition associated with arterial and venous thrombosis following vaccination with the ChAdOx1 nCoV-19 (AstraZeneca) vaccine. This report describes two cases of stroke caused by arterial and venous thromboses presenting within 28 days of receiving the AstraZeneca vaccine. The patients were otherwise young and healthy with minimal risk factors for thrombosis yet developed a rapid, ultimately fatal neurological deterioration.The patients were significantly thrombocytopenic with disproportionately raised D-dimers, both of which are widely reported in this condition. Both cases had measurable immunoglobulin G platelet factor-4 antibodies detected via enzyme-linked immunosorbent assay, similar to those described in heparin-induced thrombocytopenia.These cases illustrate that physicians should be especially mindful of VITT in the context of evolving evidence on treatment and in view of the potentially rapid and catastrophic neurological deterioration, leading to fatality despite best supportive care.

Classification of apatite structures via topological data analysis: a framework for a 'Materials Barcode' representation of structure maps.

This paper introduces the use of topological data analysis (TDA) as an unsupervised machine learning tool to uncover classification criteria in complex inorganic crystal chemistries. Using the apatite chemistry as a template, we track through the use of persistent homology the topological connectivity of input crystal chemistry descriptors on defining similarity between different stoichiometries of apatites. It is shown that TDA automatically identifies a hierarchical classification scheme within apatites based on the commonality of the number of discrete coordination polyhedra that constitute the structural building units common among the compounds. This information is presented in the form of a visualization scheme of a barcode of homology classifications, where the persistence of similarity between compounds is tracked. Unlike traditional perspectives of structure maps, this new "Materials Barcode" schema serves as an automated exploratory machine learning tool that can uncover structural associations from crystal chemistry databases, as well as to achieve a more nuanced insight into what defines similarity among homologous compounds.

Deep Learning Model for Identifying Critical Structural Motifs in Potential Endocrine Disruptors.

This paper aims to identify structural motifs within a molecule that contribute the most toward a chemical being an endocrine disruptor. We have developed a deep neural network-based toolkit toward this aim. The trained model can virtually assess a synthetic chemical's potential to be an endocrine disruptor using machine-readable molecular representation, simplified molecular input line entry system (SMILES). Our proposed toolkit is a multilabel or multioutput classification model that combines both convolution and long short-term memory (LSTM) architectures. The toolkit leverages the advantages of an active learning-based framework that combines multiple sources of data. Class activation maps (CAMs) generated from the feature-extraction layers can identify the structural alerts and the chemical environment that determines the specificity of the structural alerts.

Quantum signatures for screening metavalent solids.

The objective of this paper is to describe a new data-driven framework for computational screening and discovery of a class of materials termed "metavalent" solids. "Metavalent" solids possess characteristics that are nominally associated with metallic and covalent bonding (in terms of conductivity and coordination numbers) but are distinctly different from both because they show anomalously large response properties and a unique bond-breaking mechanism that is not observed in either covalent or metallic solids. The paper introduces the use of Hirshfeld surface analysis to provide quantum level descriptors that can be used for rapid screening of crystallographic data to identify potentially new "metavalent" solids with novel and emergent properties.

A database framework for rapid screening of structure-function relationships in PFAS chemistry.

This paper describes a database framework that enables one to rapidly explore systematics in structure-function relationships associated with new and emerging PFAS chemistries. The data framework maps high dimensional information associated with the SMILES approach of encoding molecular structure with functionality data including bioactivity and physicochemical property. This 'PFAS-Map' is a 3-dimensional unsupervised visualization tool that can automatically classify new PFAS chemistries based on current PFAS classification criteria. We provide examples on how the PFAS-Map can be utilized, including the prediction and estimation of yet unmeasured fundamental physical properties of PFAS chemistries, uncovering hierarchical characteristics in existing classification schemes, and the fusion of data from diverse sources.

Deep Learning Based Prediction of Perovskite Lattice Parameters from Hirshfeld Surface Fingerprints.

This Letter describes the use of deep learning methods on Hirshfeld surface representations of crystal structure, as an automated means of predicting lattice parameters in cubic inorganic perovskites. While Hirshfeld Surface Analysis is a well-established tool in organic crystallography, we also introduce modified computational protocols for Hirshfeld Surface Analysis tailored specifically to account for nuanced but important differences dealing with inorganic crystals. We demonstrate how two-dimensional Hirshfeld surface fingerprints can serve as a rich "database" of information encoding the complexity of relationships between chemical bonding and bond geometry characteristics of perovskites. Our results are compared with other studies on lattice parameter prediction involving both experimental and computationally derived data, and it is shown that our approach is an improvement over other reported methods. The paper concludes by discussing how this work opens new avenues for data-driven high throughput computational predictions of structure-property relationships involving complex crystal chemistries.

Data Analytics Approach for Rational Design of Nanomedicines with Programmable Drug Release.

Drug delivery vehicles can improve the functional efficacy of existing antimicrobial therapies by improving biodistribution and targeting. A critical property of such nanomedicine formulations is their ability to control the release kinetics of their payloads. The combination of (and interactions among) polymer, drug, and nanoparticle properties gives rise to nonlinear behavioral relationships and large data space. These factors complicate both first-principles modeling and screening of nanomedicine formulations. Predictive analytics may offer a more efficient approach toward the rational design of nanomedicines by identifying key descriptors and correlating them to nanoparticle release behavior. In this work, antibiotic release kinetics data were generated from polyanhydride nanoparticle formulations with varying copolymer compositions, encapsulated drug type, and drug loading. Four antibiotics, doxycycline, rifampicin, chloramphenicol, and pyrazinamide, were used. Linear manifold learning methods were used to relate drug release properties with polymer, drug, and nanoparticle properties, and key descriptors were identified that are highly correlated with release properties. However, these linear methods could not predict release behavior. Nonlinear multivariate modeling based on graph theory was then used to deconvolute the governing relationships between these properties, and predictive models were generated to rapidly screen lead nanomedicine formulations with desirable release properties with minimal nanoparticle characterization. Release kinetics predictions of two drugs containing atoms not included in the model showed good agreement with experimental results, validating the model and indicating its potential to virtually explore new polymer and drug pairs not included in the training data set. The models were shown to be robust after the inclusion of these new formulations, in that the new inclusions did not significantly change model regression. This approach provides the first step toward the development of a framework that can be used to rationally design nanomedicine formulations by selecting the appropriate carrier for a drug payload to program desirable release kinetics.

Probabilistic Assessment of Glass Forming Ability Rules for Metallic Glasses Aided by Automated Analysis of Phase Diagrams.

The use of machine learning techniques to expedite the discovery and development of new materials is an essential step towards the acceleration of a new generation of domain-specific highly functional material systems. In this paper, we use the test case of bulk metallic glasses to highlight the key issues in the field of high throughput predictions and propose a new probabilistic analysis of rules for glass forming ability using rough set theory. This approach has been applied to a broad range of binary alloy compositions in order to predict new metallic glass compositions. Our data driven approach takes into account not only a broad variety of thermodynamic, structural and kinetic based criteria, but also incorporates qualitative and descriptive attributes associated with eutectic points in phase diagrams. For the latter, we demonstrate the use of automated machine learning methods that go far beyond text recognition approaches by also being able to interpret phase diagrams. When combined with structural descriptors, this approach provides the foundations to develop a hierarchical probabilistic predication tool that can rank the feasibility of glass formation.

Wearable Intrinsically Soft, Stretchable, Flexible Devices for Memories and Computing.

A recent trend in the development of high mass consumption electron devices is towards electronic textiles (e-textiles), smart wearable devices, smart clothes, and flexible or printable electronics. Intrinsically soft, stretchable, flexible, Wearable Memories and Computing devices (WMCs) bring us closer to sci-fi scenarios, where future electronic systems are totally integrated in our everyday outfits and help us in achieving a higher comfort level, interacting for us with other digital devices such as smartphones and domotics, or with analog devices, such as our brain/peripheral nervous system. WMC will enable each of us to contribute to open and big data systems as individual nodes, providing real-time information about physical and environmental parameters (including air pollution monitoring, sound and light pollution, chemical or radioactive fallout alert, network availability, and so on). Furthermore, WMC could be directly connected to human brain and enable extremely fast operation and unprecedented interface complexity, directly mapping the continuous states available to biological systems. This review focuses on recent advances in nanotechnology and materials science and pays particular attention to any result and promising technology to enable intrinsically soft, stretchable, flexible WMC.

Functionalization promotes pathogen-mimicking characteristics of polyanhydride nanoparticle adjuvants.

Rational design of adjuvants and delivery systems will promote development of next-generation vaccines to control emerging and re-emerging diseases. To accomplish this, understanding the immune-enhancing properties of new adjuvants relative to those induced by natural infections can help with the development of pathogen-mimicking materials that will effectively initiate innate immune signaling cascades. In this work, the surfaces of polyanhydride nanoparticles composed of sebacic acid (SA) and 1,6-bis(p-carboxyphenoxy) hexane were decorated with an ethylene diamine spacer partially modified with either a glycolic acid linker or an α-1,2-linked di-mannopyranoside (di-mannose) to confer "pathogen-like" properties and enhance adjuvanticity. Co-incubation of linker-modified nanoparticles with dendritic cells (DCs) elicited significant increases in surface expression of MHC I, MHC II, CD86, and CD40, and enhanced secretion of IL-6, IL-12p40, and TNF-α. An 800% increase in uptake of ethylene-diamine-spaced, linker and di-mannose functionalized polyanhydride nanoparticles was also observed. Together, our data showed that linker-functionalized polyanhydride nanoparticles demonstrate similar patterns of uptake, intracellular trafficking, particle persistence, and innate activation as did DCs exposed to Yersinia pestis or Escherichia coli. These results set the stage for rational selection of adjuvant chemistries to induce pathogen-mimicking immune responses. © 2017 Wiley Periodicals, Inc. J Biomed Mater Res Part A: 105A: 2762-2771, 2017.

Silver nanoparticle ink technology: state of the art.

Printed electronics will bring to the consumer level great breakthroughs and unique products in the near future, shifting the usual paradigm of electronic devices and circuit boards from hard boxes and rigid sheets into flexible thin layers and bringing disposable electronics, smart tags, and so on. The most promising tool to achieve the target depends upon the availability of nanotechnology-based functional inks. A certain delay in the innovation-transfer process to the market is now being observed. Nevertheless, the most widely diffused product, settled technology, and the highest sales volumes are related to the silver nanoparticle-based ink market, representing the best example of commercial nanotechnology today. This is a compact review on synthesis routes, main properties, and practical applications.

Combination Nanovaccine Demonstrates Synergistic Enhancement in Efficacy against Influenza.

H5N1 influenza virus has the potential to become a significant global health threat, and next generation vaccine technologies are needed. In this work, the combined efficacy of two nanoadjuvant platforms (polyanhydride nanoparticles and pentablock copolymer-based hydrogels) to induce protective immunity against H5N1 influenza virus was examined. Mice received two subcutaneous vaccinations (day 0 and 21) containing 10 µg of H5 hemagglutinin trimer alone or in combination with the nanovaccine platforms. Nanovaccine immunization induced high neutralizing antibody titers that were sustained through 70 days postimmunization. Finally, mice were intranasally challenged with A/H5N1 VNH5N1-PR8CDC-RG virus and monitored for 14 days. Animals receiving the combination nanovaccine had lower viral loads in the lung and weight loss after challenge in comparison to animals vaccinated with each platform alone. These data demonstrate the synergy between polyanhydride nanoparticles and pentablock copolymer-based hydrogels as adjuvants in the design of a more efficacious influenza vaccine.

Mapping Chemical Selection Pathways for Designing Multicomponent Alloys: an informatics framework for materials design.

A data driven methodology is developed for tracking the collective influence of the multiple attributes of alloying elements on both thermodynamic and mechanical properties of metal alloys. Cobalt-based superalloys are used as a template to demonstrate the approach. By mapping the high dimensional nature of the systematics of elemental data embedded in the periodic table into the form of a network graph, one can guide targeted first principles calculations that identify the influence of specific elements on phase stability, crystal structure and elastic properties. This provides a fundamentally new means to rapidly identify new stable alloy chemistries with enhanced high temperature properties. The resulting visualization scheme exhibits the grouping and proximity of elements based on their impact on the properties of intermetallic alloys. Unlike the periodic table however, the distance between neighboring elements uncovers relationships in a complex high dimensional information space that would not have been easily seen otherwise. The predictions of the methodology are found to be consistent with reported experimental and theoretical studies. The informatics based methodology presented in this study can be generalized to a framework for data analysis and knowledge discovery that can be applied to many material systems and recreated for different design objectives.

Mining information from atom probe data.

Whilst atom probe tomography (APT) is a powerful technique with the capacity to gather information containing hundreds of millions of atoms from a single specimen, the ability to effectively use this information creates significant challenges. The main technological bottleneck lies in handling the extremely large amounts of data on spatial-chemical correlations, as well as developing new quantitative computational foundations for image reconstruction that target critical and transformative problems in materials science. The power to explore materials at the atomic scale with the extraordinary level of sensitivity of detection offered by atom probe tomography has not been not fully harnessed due to the challenges of dealing with missing, sparse and often noisy data. Hence there is a profound need to couple the analytical tools to deal with the data challenges with the experimental issues associated with this instrument. In this paper we provide a summary of some key issues associated with the challenges, and solutions to extract or "mine" fundamental materials science information from that data.

Extracting features buried within high density atom probe point cloud data through simplicial homology.

Feature extraction from Atom Probe Tomography (APT) data is usually performed by repeatedly delineating iso-concentration surfaces of a chemical component of the sample material at different values of concentration threshold, until the user visually determines a satisfactory result in line with prior knowledge. However, this approach allows for important features, buried within the sample, to be visually obscured by the high density and volume (~10(7) atoms) of APT data. This work provides a data driven methodology to objectively determine the appropriate concentration threshold for classifying different phases, such as precipitates, by mapping the topology of the APT data set using a concept from algebraic topology termed persistent simplicial homology. A case study of Sc precipitates in an Al-Mg-Sc alloy is presented demonstrating the power of this technique to capture features, such as precise demarcation of Sc clusters and Al segregation at the cluster boundaries, not easily available by routine visual adjustment.

Automated voxelization of 3D atom probe data through kernel density estimation.

Identifying nanoscale chemical features from atom probe tomography (APT) data routinely involves adjustment of voxel size as an input parameter, through visual supervision, making the final outcome user dependent, reliant on heuristic knowledge and potentially prone to error. This work utilizes Kernel density estimators to select an optimal voxel size in an unsupervised manner to perform feature selection, in particular targeting resolution of interfacial features and chemistries. The capability of this approach is demonstrated through analysis of the γ / γ' interface in a Ni-Al-Cr superalloy.