Improving predictions of compound amenability for liquid chromatography-mass spectrometry to enhance non-targeted analysis.

Mass-spectrometry-based non-targeted analysis (NTA), in which mass spectrometric signals are assigned chemical identities based on a systematic collation of evidence, is a growing area of interest for toxicological risk assessment. Successful NTA results in better identification of potentially hazardous pollutants within the environment, facilitating the development of targeted analytical strategies to best characterize risks to human and ecological health. A supporting component of the NTA process involves assessing whether suspected chemicals are amenable to the mass spectrometric method, which is necessary in order to assign an observed signal to the chemical structure. Prior work from this group involved the development of a random forest model for predicting the amenability of 5517 unique chemical structures to liquid chromatography-mass spectrometry (LC-MS). This work improves the interpretability of the group's prior model of the same endpoint, as well as integrating 1348 more data points across negative and positive ionization modes. We enhance interpretability by feature engineering, a machine learning practice that reduces the input dimensionality while attempting to preserve performance statistics. We emphasize the importance of interpretable machine learning models within the context of building confidence in NTA identification. The novel data were curated by the labeling of compounds as amenable or unamenable by expert curators, resulting in an enhanced set of chemical compounds to expand the applicability domain of the prior model. The balanced accuracy benchmark of the newly developed model is comparable to performance previously reported (mean CV BA is 0.84 vs. 0.82 in positive mode, and 0.85 vs. 0.82 in negative mode), while on a novel external set, derived from this work's data, the Matthews correlation coefficients (MCC) for the novel models are 0.66 and 0.68 for positive and negative mode, respectively. Our group's prior published models scored MCC of 0.55 and 0.54 on the same external sets. This demonstrates appreciable improvement over the chemical space captured by the expanded dataset. This work forms part of our ongoing efforts to develop models with higher interpretability and higher performance to support NTA efforts.

Free and open-source QSAR-ready workflow for automated standardization of chemical structures in support of QSAR modeling.

The rapid increase of publicly available chemical structures and associated experimental data presents a valuable opportunity to build robust QSAR models for applications in different fields. However, the common concern is the quality of both the chemical structure information and associated experimental data. This is especially true when those data are collected from multiple sources as chemical substance mappings can contain many duplicate structures and molecular inconsistencies. Such issues can impact the resulting molecular descriptors and their mappings to experimental data and, subsequently, the quality of the derived models in terms of accuracy, repeatability, and reliability. Herein we describe the development of an automated workflow to standardize chemical structures according to a set of standard rules and generate two and/or three-dimensional "QSAR-ready" forms prior to the calculation of molecular descriptors. The workflow was designed in the KNIME workflow environment and consists of three high-level steps. First, a structure encoding is read, and then the resulting in-memory representation is cross-referenced with any existing identifiers for consistency. Finally, the structure is standardized using a series of operations including desalting, stripping of stereochemistry (for two-dimensional structures), standardization of tautomers and nitro groups, valence correction, neutralization when possible, and then removal of duplicates. This workflow was initially developed to support collaborative modeling QSAR projects to ensure consistency of the results from the different participants. It was then updated and generalized for other modeling applications. This included modification of the "QSAR-ready" workflow to generate "MS-ready structures" to support the generation of substance mappings and searches for software applications related to non-targeted analysis mass spectrometry. Both QSAR and MS-ready workflows are freely available in KNIME, via standalone versions on GitHub, and as docker container resources for the scientific community. Scientific contribution: This work pioneers an automated workflow in KNIME, systematically standardizing chemical structures to ensure their readiness for QSAR modeling and broader scientific applications. By addressing data quality concerns through desalting, stereochemistry stripping, and normalization, it optimizes molecular descriptors' accuracy and reliability. The freely available resources in KNIME, GitHub, and docker containers democratize access, benefiting collaborative research and advancing diverse modeling endeavors in chemistry and mass spectrometry.

Transparency in Modeling through Careful Application of OECD's QSAR/QSPR Principles via a Curated Water Solubility Data Set.

The need for careful assembly, training, and validation of quantitative structure-activity/property models (QSAR/QSPR) is more significant than ever as data sets become larger and sophisticated machine learning tools become increasingly ubiquitous and accessible to the scientific community. Regulatory agencies such as the United States Environmental Protection Agency must carefully scrutinize each aspect of a resulting QSAR/QSPR model to determine its potential use in environmental exposure and hazard assessment. Herein, we revisit the goals of the Organisation for Economic Cooperation and Development (OECD) in our application and discuss the validation principles for structure-activity models. We apply these principles to a model for predicting water solubility of organic compounds derived using random forest regression, a common machine learning approach in the QSA/PR literature. Using public sources, we carefully assembled and curated a data set consisting of 10,200 unique chemical structures with associated water solubility measurements. This data set was then used as a focal narrative to methodically consider the OECD's QSA/PR principles and how they can be applied to random forests. Despite some expert, mechanistically informed supervision of descriptor selection to enhance model interpretability, we achieved a model of water solubility with comparable performance to previously published models (5-fold cross validated performance 0.81 R2 and 0.98 RMSE). We hope this work will catalyze a necessary conversation around the importance of cautiously modernizing and explicitly leveraging OECD principles while pursuing state-of-the-art machine learning approaches to derive QSA/PR models suitable for regulatory consideration.

Exploring chemical space in non-targeted analysis: a proposed ChemSpace tool.

Non-targeted analysis (NTA) using high-resolution mass spectrometry allows scientists to detect and identify a broad range of compounds in diverse matrices for monitoring exposure and toxicological evaluation without a priori chemical knowledge. NTA methods present an opportunity to describe the constituents of a sample across a multidimensional swath of chemical properties, referred to as "chemical space." Understanding and communicating which region of chemical space is extractable and detectable by an NTA workflow, however, remains challenging and non-standardized. For example, many sample processing and data analysis steps influence the types of chemicals that can be detected and identified. Accordingly, it is challenging to assess whether analyte non-detection in an NTA study indicates true absence in a sample (above a detection limit) or is a false negative driven by workflow limitations. Here, we describe the need for accessible approaches that enable chemical space mapping in NTA studies, propose a tool to address this need, and highlight the different ways in which it could be implemented in NTA workflows. We identify a suite of existing predictive and analytical tools that can be used in combination to generate scores that describe the likelihood a compound will be detected and identified by a given NTA workflow based on the predicted chemical space of that workflow. Higher scores correspond to a higher likelihood of compound detection and identification in a given workflow (based on sample extraction, data acquisition, and data analysis parameters). Lower scores indicate a lower probability of detection, even if the compound is truly present in the samples of interest. Understanding the constraints of NTA workflows can be useful for stakeholders when results from NTA studies are used in real-world applications and for NTA researchers working to improve their workflow performance. The hypothetical ChemSpaceTool suggested herein could be used in both a prospective and retrospective sense. Prospectively, the tool can be used to further curate screening libraries and set identification thresholds. Retrospectively, false detections can be filtered by the plausibility of the compound identification by the selected NTA method, increasing the confidence of unknown identifications. Lastly, this work highlights the chemometric needs to make such a tool robust and usable across a wide range of NTA disciplines and invites others who are working on various models to participate in the development of the ChemSpaceTool. Ultimately, the development of a chemical space mapping tool strives to enable further standardization of NTA by improving method transparency and communication around false detection rates, thus allowing for more direct method comparisons between studies and improved reproducibility. This, in turn, is expected to promote further widespread applications of NTA beyond research-oriented settings.

Assembly and Curation of Lists of Per- and Polyfluoroalkyl Substances (PFAS) to Support Environmental Science Research.

Per- and polyfluoroalkyl substances (PFAS) are a class of man-made chemicals of global concern for many health and regulatory agencies due to their widespread use and persistence in the environment (in soil, air, and water), bioaccumulation, and toxicity. This concern has catalyzed a need to aggregate data to support research efforts that can, in turn, inform regulatory and statutory actions. An ongoing challenge regarding PFAS has been the shifting definition of what qualifies a substance to be a member of the PFAS class. There is no single definition for a PFAS, but various attempts have been made to utilize substructural definitions that either encompass broad working scopes or satisfy narrower regulatory guidelines. Depending on the size and specificity of PFAS substructural filters applied to the U.S. Environmental Protection Agency (EPA) DSSTox database, currently exceeding 900,000 unique substances, PFAS substructure-defined space can span hundreds to tens of thousands of compounds. This manuscript reports on the curation of PFAS chemicals and assembly of lists that have been made publicly available to the community via the EPA's CompTox Chemicals Dashboard. Creation of these PFAS lists required the harvesting of data from EPA and online databases, peer-reviewed publications, and regulatory documents. These data have been extracted and manually curated, annotated with structures, and made available to the community in the form of lists defined by structure filters, as well as lists comprising non-structurable PFAS, such as polymers and complex mixtures. These lists, along with their associated linkages to predicted and measured data, are fueling PFAS research efforts within the EPA and are serving as a valuable resource to the international scientific community.

Uncertainty estimation strategies for quantitative non-targeted analysis.

Non-targeted analysis (NTA) methods are widely used for chemical discovery but seldom employed for quantitation due to a lack of robust methods to estimate chemical concentrations with confidence limits. Herein, we present and evaluate new statistical methods for quantitative NTA (qNTA) using high-resolution mass spectrometry (HRMS) data from EPA's Non-Targeted Analysis Collaborative Trial (ENTACT). Experimental intensities of ENTACT analytes were observed at multiple concentrations using a semi-automated NTA workflow. Chemical concentrations and corresponding confidence limits were first estimated using traditional calibration curves. Two qNTA estimation methods were then implemented using experimental response factor (RF) data (where RF = intensity/concentration). The bounded response factor method used a non-parametric bootstrap procedure to estimate select quantiles of training set RF distributions. Quantile estimates then were applied to test set HRMS intensities to inversely estimate concentrations with confidence limits. The ionization efficiency estimation method restricted the distribution of likely RFs for each analyte using ionization efficiency predictions. Given the intended future use for chemical risk characterization, predicted upper confidence limits (protective values) were compared to known chemical concentrations. Using traditional calibration curves, 95% of upper confidence limits were within ~tenfold of the true concentrations. The error increased to ~60-fold (ESI+) and ~120-fold (ESI-) for the ionization efficiency estimation method and to ~150-fold (ESI+) and ~130-fold (ESI-) for the bounded response factor method. This work demonstrates successful implementation of confidence limit estimation strategies to support qNTA studies and marks a crucial step towards translating NTA data in a risk-based context.

Predicting compound amenability with liquid chromatography-mass spectrometry to improve non-targeted analysis.

With the increasing availability of high-resolution mass spectrometers, suspect screening and non-targeted analysis are becoming popular compound identification tools for environmental researchers. Samples of interest often contain a large (unknown) number of chemicals spanning the detectable mass range of the instrument. In an effort to separate these chemicals prior to injection into the mass spectrometer, a chromatography method is often utilized. There are numerous types of gas and liquid chromatographs that can be coupled to commercially available mass spectrometers. Depending on the type of instrument used for analysis, the researcher is likely to observe a different subset of compounds based on the amenability of those chemicals to the selected experimental techniques and equipment. It would be advantageous if this subset of chemicals could be predicted prior to conducting the experiment, in order to minimize potential false-positive and false-negative identifications. In this work, we utilize experimental datasets to predict the amenability of chemical compounds to detection with liquid chromatography-electrospray ionization-mass spectrometry (LC-ESI-MS). The assembled dataset totals 5517 unique chemicals either explicitly detected or not detected with LC-ESI-MS. The resulting detected/not-detected matrix has been modeled using specific molecular descriptors to predict which chemicals are amenable to LC-ESI-MS, and to which form(s) of ionization. Random forest models, including a measure of the applicability domain of the model for both positive and negative modes of the electrospray ionization source, were successfully developed. The outcome of this work will help to inform future suspect screening and non-targeted analyses of chemicals by better defining the potential LC-ESI-MS detectable chemical landscape of interest.

Chemical Characterization of Recycled Consumer Products Using Suspect Screening Analysis.

Recycled materials are found in many consumer products as part of a circular economy; however, the chemical content of recycled products is generally uncharacterized. A suspect screening analysis using two-dimensional gas chromatography time-of-flight mass spectrometry (GC × GC-TOFMS) was applied to 210 products (154 recycled, 56 virgin) across seven categories. Chemicals in products were tentatively identified using a standard spectral library or confirmed using chemical standards. A total of 918 probable chemical structures identified (112 of which were confirmed) in recycled materials versus 587 (110 confirmed) in virgin materials. Identified chemicals were characterized in terms of their functional use and structural class. Recycled paper products and construction materials contained greater numbers of chemicals than virgin products; 733 identified chemicals had greater occurrence in recycled compared to virgin materials. Products made from recycled materials contained greater numbers of fragrances, flame retardants, solvents, biocides, and dyes. The results were clustered to identify groups of chemicals potentially associated with unique chemical sources, and identified chemicals were prioritized for further study using high-throughput hazard and exposure information. While occurrence is not necessarily indicative of risk, these results can be used to inform the expansion of existing models or identify exposure pathways currently neglected in exposure assessments.

Using the US EPA CompTox Chemicals Dashboard to interpret targeted and non-targeted GC-MS analyses from human breath and other biological media.

The U.S. EPA CompTox Chemicals Dashboard is a freely available web-based application providing access to chemistry, toxicity, and exposure data for ∼900 000 chemicals. Data, search functionality, and prediction models within the Dashboard can help identify chemicals found in environmental analyses and human biomonitoring. It was designed to deliver data generated to support computational toxicology to reduce chemical testing on animals and provide access to new approach methodologies including prediction models. The inclusion of mass and formula-based searches, together with relevant ranking approaches, allows for the identification and prioritization of exogenous (environmental) chemicals from high resolution mass spectrometry in need of further evaluation. The Dashboard includes chemicals that can be detected by liquid chromatography, gas chromatography-mass spectrometry (GC-MS) and direct-MS analyses, and chemical lists have been added that highlight breath-borne volatile and semi-volatile organic compounds. The Dashboard can be searched using various chemical identifiers (e.g. chemical synonyms, CASRN and InChIKeys), chemical formula, MS-ready formulae monoisotopic mass, consumer product categories and assays/genes associated with high-throughput screening data. An integrated search at a chemical level performs searches against PubMed to identify relevant published literature. This article describes specific procedures using the Dashboard as a first-stop tool for exploring both targeted and non-targeted results from GC-MS analyses of chemicals found in breath, exhaled breath condensate, and associated aerosols.

Enabling High-Throughput Searches for Multiple Chemical Data Using the U.S.-EPA CompTox Chemicals Dashboard.

The core goal of cheminformatics is to efficiently store robust and accurate chemical information and make it accessible for drug discovery, environmental analysis, and the development of prediction models including quantitative structure-activity relationships (QSAR). The U.S. Environmental Protection Agency (EPA) has developed a web-based application, the CompTox Chemicals Dashboard, which provides access to a compilation of data generated within the agency and sourced from public databases and literature and to utilities for real-time QSAR prediction and chemical read-across. While the vast majority of online tools only allow interrogation of chemicals one at a time, the Dashboard provides a batch search feature that allows for the sourcing of data based on thousands of chemical inputs at one time, by chemical identifier (e.g., names, Chemical Abstract Service registry numbers, or InChIKeys), or by mass or molecular formulas. Chemical information that can then be sourced via the batch search includes chemical identifiers and structures; intrinsic, physicochemical and fate and transport properties; in vitro and in vivo toxicity data; and the presence in environmentally relevant lists. We outline how to use the batch search feature and provide an overview regarding the type of information that can be sourced by considering a series of typical-use questions.

Revisiting Five Years of CASMI Contests with EPA Identification Tools.

Software applications for high resolution mass spectrometry (HRMS)-based non-targeted analysis (NTA) continue to enhance chemical identification capabilities. Given the variety of available applications, determining the most fit-for-purpose tools and workflows can be difficult. The Critical Assessment of Small Molecule Identification (CASMI) contests were initiated in 2012 to provide a means to evaluate compound identification tools on a standardized set of blinded tandem mass spectrometry (MS/MS) data. Five CASMI contests have resulted in recommendations, publications, and invaluable datasets for practitioners of HRMS-based screening studies. The US Environmental Protection Agency's (EPA) CompTox Chemicals Dashboard is now recognized as a valuable resource for compound identification in NTA studies. However, this application was too new and immature in functionality to participate in the five previous CASMI contests. In this work, we performed compound identification on all five CASMI contest datasets using Dashboard tools and data in order to critically evaluate Dashboard performance relative to that of other applications. CASMI data was accessed via the CASMI webpage and processed for use in our spectral matching and identification workflow. Relative to applications used by former contest participants, our tools, data, and workflow performed well, placing more challenge compounds in the top five of ranked candidates than did the winners of three contest years and tying in a fourth. In addition, we conducted an in-depth review of the CASMI structure sets and made these reviewed sets available via the Dashboard. Our results suggest that Dashboard data and tools would enhance chemical identification capabilities for practitioners of HRMS-based NTA.

Telemedicine: Is it really worth it? A perspective from evidence and experience.

BACKGROUND:  Although the formal evidence base is equivocal, practical experience suggests that implementations of technology that support telemedicine initiatives can result in improved patient outcomes, better patient and carer experience and reduced expenditure. OBJECTIVE: To answer the questions "Is an investment in telemedicine worth it?" and "How do I make a telemedicine implementation work?" METHODS:  Summary of systematic review evidence and an illustrative case study. Discussion of implications for industry and policy. RESULTS: Realisation of telemedicine benefits is much less to do with the technology itself and much more around the context of the implementing organisation and its ability to implement. CONCLUSION:  We recommend that local organisations consider deployment of telemedicine initiatives but with a greater awareness of the growing body of implementation best practice. We also recommend, for the NHS, that the centre takes a greater role in the collation and dissemination of best practice to support successful implementations of telemedicine and other health informatics initiatives.

The use of kiosk technology in general practice.

We have attempted to quantify the benefit to both patients and professional users from introducing 'self-service' kiosks into the waiting rooms of general practitioners (GPs). We conducted a series of interviews with practice managers, GPs and patients in practices where kiosks had been used. We then estimated the time savings and other benefits. There were significant time savings both for nursing and GP staff, and similar time savings for patients. We estimate that the total time savings would be equivalent to one health-care assistant for a practice size of 10-12,000. Practice income was enhanced as a result of the improved information gathering made possible by the kiosks. These effects resulted in improved patient outcomes. The key to successful implementation appears to be strong support, particularly from reception staff to encourage and facilitate the use of kiosks. To maximise the benefits, GPs and nurses must be prepared to change their ways of working.

Glaucoma with the oculocerebrorenal syndrome of Lowe.

PURPOSE: To further describe the glaucoma with the oculocerebrorenal syndrome of Lowe (OCRL) including the responsible filtration angle abnormalities and response to treatment. METHODS: The scientific literature regarding the glaucoma associated with OCRL from 1952, when the first report of the syndrome appeared, to the present was reviewed. The medical records of 7 patients with OCRL were studied. The occurrence of glaucoma, corneal changes secondary to glaucoma, gonioscopic abnormalities, iris features, and response to glaucoma surgery were recorded. RESULTS: Signs of glaucoma are defining abnormalities leading to recognition of OCRL. The OCRL medical literature reports the frequency of glaucoma, secondary clinical signs of increased intraocular pressure (IOP), and results of glaucoma surgery, but little information related to the responsible filtration angle abnormalities. Glaucoma was present in 71% (5 of 7) of patients studied, and was recognized in infancy in 9 of their 10 eyes. Gonioscopy was performed in 6 OCRL patients and revealed the constant presence of open angles, primary filtration angle anomalies, and defects considered secondary to previous infantile lens extractions. The observed anomalies were anterior insertion of the iris, narrowing of the ciliary body band, and decreased visibility of the scleral spur. The angle defects felt to be acquired following lens surgery included a more anterior insertion of the iris on to the trabecular meshwork, pigment dusting of the angle tissues, and iris synechial abnormalities related to the surgery. Bilateral goniotomies were unsuccessful in 4 of 4 patients. The fundi of 6 of 7 patients showed normal optic disc development with variable abnormalities secondary to glaucoma, and normal retinal vessels. Minimal evidence of macular development was observed in a single patient from 1 month of age to his most recent examination at 14 months of age. CONCLUSION: A primary X-linked infantile glaucoma is a defining and frequent component of OCRL and is secondary to expression of a primary filtration angle anomaly. Goniotomy was unsuccessful in all (8) operated eyes. The adverse effects of cataract surgery on the filtration angle structures may influence the results of goniotomy surgery by superimposing a secondary aphakic glaucoma component that may explain the need for alternative glaucoma surgery.