The Next Frontier of Environmental Unknowns: Substances of Unknown or Variable Composition, Complex Reaction Products, or Biological Materials (UVCBs).

Substances of unknown or variable composition, complex reaction products, or biological materials (UVCBs) are over 70 000 "complex" chemical mixtures produced and used at significant levels worldwide. Due to their unknown or variable composition, applying chemical assessments originally developed for individual compounds to UVCBs is challenging, which impedes sound management of these substances. Across the analytical sciences, toxicology, cheminformatics, and regulatory practice, new approaches addressing specific aspects of UVCB assessment are being developed, albeit in a fragmented manner. This review attempts to convey the "big picture" of the state of the art in dealing with UVCBs by holistically examining UVCB characterization and chemical identity representation, as well as hazard, exposure, and risk assessment. Overall, information gaps on chemical identities underpin the fundamental challenges concerning UVCBs, and better reporting and substance characterization efforts are needed to support subsequent chemical assessments. To this end, an information level scheme for improved UVCB data collection and management within databases is proposed. The development of UVCB testing shows early progress, in line with three main methods: whole substance, known constituents, and fraction profiling. For toxicity assessment, one option is a whole-mixture testing approach. If the identities of (many) constituents are known, grouping, read across, and mixture toxicity modeling represent complementary approaches to overcome data gaps in toxicity assessment. This review highlights continued needs for concerted efforts from all stakeholders to ensure proper assessment and sound management of UVCBs.

Using Machine Learning to Parse Chemical Mixture Descriptions.

Chemical mixtures have recently come to the attention of open standards and data structures for capturing machine-readable descriptions for informatics uses. At the present time, essentially all transmission of information about mixtures is done using short text descriptions that are readable only by trained scientists, and there are no accessible repositories of marked-up mixture data. We have designed a machine learning tool that can interpret mixture descriptions and upgrade them to the high-level Mixfile format, which can in turn be used to generate Mixtures InChI notation. The interpretation achieves a high success rate and can be used at scale to markup large catalogs and inventories, with some expert checking to catch edge cases. The training data that was accumulated during the project is made openly available, along with previously released mixture editing tools and utilities.

CATMoS: Collaborative Acute Toxicity Modeling Suite.

BACKGROUND: Humans are exposed to tens of thousands of chemical substances that need to be assessed for their potential toxicity. Acute systemic toxicity testing serves as the basis for regulatory hazard classification, labeling, and risk management. However, it is cost- and time-prohibitive to evaluate all new and existing chemicals using traditional rodent acute toxicity tests. In silico models built using existing data facilitate rapid acute toxicity predictions without using animals. OBJECTIVES: The U.S. Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) Acute Toxicity Workgroup organized an international collaboration to develop in silico models for predicting acute oral toxicity based on five different end points: Lethal Dose 50 (LD50 value, U.S. Environmental Protection Agency hazard (four) categories, Globally Harmonized System for Classification and Labeling hazard (five) categories, very toxic chemicals [LD50 (LD50≤50mg/kg)], and nontoxic chemicals (LD50>2,000mg/kg). METHODS: An acute oral toxicity data inventory for 11,992 chemicals was compiled, split into training and evaluation sets, and made available to 35 participating international research groups that submitted a total of 139 predictive models. Predictions that fell within the applicability domains of the submitted models were evaluated using external validation sets. These were then combined into consensus models to leverage strengths of individual approaches. RESULTS: The resulting consensus predictions, which leverage the collective strengths of each individual model, form the Collaborative Acute Toxicity Modeling Suite (CATMoS). CATMoS demonstrated high performance in terms of accuracy and robustness when compared with in vivo results. DISCUSSION: CATMoS is being evaluated by regulatory agencies for its utility and applicability as a potential replacement for in vivo rat acute oral toxicity studies. CATMoS predictions for more than 800,000 chemicals have been made available via the National Toxicology Program's Integrated Chemical Environment tools and data sets (ice.ntp.niehs.nih.gov). The models are also implemented in a free, standalone, open-source tool, OPERA, which allows predictions of new and untested chemicals to be made. https://doi.org/10.1289/EHP8495.

Capturing mixture composition: an open machine-readable format for representing mixed substances.

We describe a file format that is designed to represent mixtures of compounds in a way that is fully machine readable. This Mixfile format is intended to fill the same role for substances that are composed of multiple components as the venerable Molfile does for specifying individual structures. This much needed datastructure is intended to replace current practices for communicating information about mixtures, which usually relies on human-readable text descriptions, drawing several species within a single molecular diagram, or mutually incompatible ad hoc solutions. We describe an open source software application for editing mixture files, which can also be used as web-ready tools for manipulating the file format. We also present a corpus of mixture examples, which we have extracted from collections of text-based descriptions. Furthermore, we present an early look at the proposed IUPAC Mixtures InChI specification, instances of which can be automatically generated using the Mixfile format as a precursor.

Exploiting machine learning for end-to-end drug discovery and development.

A variety of machine learning methods such as naive Bayesian, support vector machines and more recently deep neural networks are demonstrating their utility for drug discovery and development. These leverage the generally bigger datasets created from high-throughput screening data and allow prediction of bioactivities for targets and molecular properties with increased levels of accuracy. We have only just begun to exploit the potential of these techniques but they may already be fundamentally changing the research process for identifying new molecules and/or repurposing old drugs. The integrated application of such machine learning models for end-to-end (E2E) application is broadly relevant and has considerable implications for developing future therapies and their targeting.

High-throughput screening and Bayesian machine learning for copper-dependent inhibitors of Staphylococcus aureus.

One potential source of new antibacterials is through probing existing chemical libraries for copper-dependent inhibitors (CDIs), i.e., molecules with antibiotic activity only in the presence of copper. Recently, our group demonstrated that previously unknown staphylococcal CDIs were frequently present in a small pilot screen. Here, we report the outcome of a larger industrial anti-staphylococcal screen consisting of 40 771 compounds assayed in parallel, both in standard and in copper-supplemented media. Ultimately, 483 had confirmed copper-dependent IC50 values under 50 µM. Sphere-exclusion clustering revealed that these hits were largely dominated by sulfur-containing motifs, including benzimidazole-2-thiones, thiadiazines, thiazoline formamides, triazino-benzimidazoles, and pyridinyl thieno-pyrimidines. Structure-activity relationship analysis of the pyridinyl thieno-pyrimidines generated multiple improved CDIs, with activity likely dependent on ligand/ion coordination. Molecular fingerprint-based Bayesian classification models were built using Discovery Studio and Assay Central, a new platform for sharing and distributing cheminformatic models in a portable format, based on open-source tools. Finally, we used the latter model to evaluate a library of FDA-approved drugs for copper-dependent activity in silico. Two anti-helminths, albendazole and thiabendazole, scored highly and are known to coordinate copper ions, further validating the model's applicability.

Ebola Virus Bayesian Machine Learning Models Enable New in Vitro Leads.

We have previously described the first Bayesian machine learning models from FDA-approved drug screens, for identifying compounds active against the Ebola virus (EBOV). These models led to the identification of three active molecules in vitro: tilorone, pyronaridine, and quinacrine. A follow-up study demonstrated that one of these compounds, tilorone, has 100% in vivo efficacy in mice infected with mouse-adapted EBOV at 30 mg/kg/day intraperitoneal. This suggested that we can learn from the published data on EBOV inhibition and use it to select new compounds for testing that are active in vivo. We used these previously built Bayesian machine learning EBOV models alongside our chemical insights for the selection of 12 molecules, absent from the training set, to test for in vitro EBOV inhibition. Nine molecules were directly selected using the model, and eight of these molecules possessed a promising in vitro activity (EC50 < 15 µM). Three further compounds were selected for an in vitro evaluation because they were antimalarials, and compounds of this class like pyronaridine and quinacrine have previously been shown to inhibit EBOV. We identified the antimalarial drug arterolane (IC50 = 4.53 µM) and the anticancer clinical candidate lucanthone (IC50 = 3.27 µM) as novel compounds that have EBOV inhibitory activity in HeLa cells and generally lack cytotoxicity. This work provides further validation for using machine learning and medicinal chemistry expertize to prioritize compounds for testing in vitro prior to more costly in vivo tests. These studies provide further corroboration of this strategy and suggest that it can likely be applied to other pathogens in the future.

Multiple Machine Learning Comparisons of HIV Cell-based and Reverse Transcriptase Data Sets.

The human immunodeficiency virus (HIV) causes over a million deaths every year and has a huge economic impact in many countries. The first class of drugs approved were nucleoside reverse transcriptase inhibitors. A newer generation of reverse transcriptase inhibitors have become susceptible to drug resistant strains of HIV, and hence, alternatives are urgently needed. We have recently pioneered the use of Bayesian machine learning to generate models with public data to identify new compounds for testing against different disease targets. The current study has used the NIAID ChemDB HIV, Opportunistic Infection and Tuberculosis Therapeutics Database for machine learning studies. We curated and cleaned data from HIV-1 wild-type cell-based and reverse transcriptase (RT) DNA polymerase inhibition assays. Compounds from this database with ≤1 µM HIV-1 RT DNA polymerase activity inhibition and cell-based HIV-1 inhibition are correlated (Pearson r = 0.44, n = 1137, p < 0.0001). Models were trained using multiple machine learning approaches (Bernoulli Naive Bayes, AdaBoost Decision Tree, Random Forest, support vector classification, k-Nearest Neighbors, and deep neural networks as well as consensus approaches) and then their predictive abilities were compared. Our comparison of different machine learning methods demonstrated that support vector classification, deep learning, and a consensus were generally comparable and not significantly different from each other using 5-fold cross validation and using 24 training and test set combinations. This study demonstrates findings in line with our previous studies for various targets that training and testing with multiple data sets does not demonstrate a significant difference between support vector machine and deep neural networks.

Comparing Multiple Machine Learning Algorithms and Metrics for Estrogen Receptor Binding Prediction.

Many chemicals that disrupt endocrine function have been linked to a variety of adverse biological outcomes. However, screening for endocrine disruption using in vitro or in vivo approaches is costly and time-consuming. Computational methods, e.g., quantitative structure-activity relationship models, have become more reliable due to bigger training sets, increased computing power, and advanced machine learning algorithms, such as multilayered artificial neural networks. Machine learning models can be used to predict compounds for endocrine disrupting capabilities, such as binding to the estrogen receptor (ER), and allow for prioritization and further testing. In this work, an exhaustive comparison of multiple machine learning algorithms, chemical spaces, and evaluation metrics for ER binding was performed on public data sets curated using in-house cheminformatics software (Assay Central). Chemical features utilized in modeling consisted of binary fingerprints (ECFP6, FCFP6, ToxPrint, or MACCS keys) and continuous molecular descriptors from RDKit. Each feature set was subjected to classic machine learning algorithms (Bernoulli Naive Bayes, AdaBoost Decision Tree, Random Forest, Support Vector Machine) and Deep Neural Networks (DNN). Models were evaluated using a variety of metrics: recall, precision, F1-score, accuracy, area under the receiver operating characteristic curve, Cohen's Kappa, and Matthews correlation coefficient. For predicting compounds within the training set, DNN has an accuracy higher than that of other methods; however, in 5-fold cross validation and external test set predictions, DNN and most classic machine learning models perform similarly regardless of the data set or molecular descriptors used. We have also used the rank normalized scores as a performance-criteria for each machine learning method, and Random Forest performed best on the validation set when ranked by metric or by data sets. These results suggest classic machine learning algorithms may be sufficient to develop high quality predictive models of ER activity.

Assessment of Substrate-Dependent Ligand Interactions at the Organic Cation Transporter OCT2 Using Six Model Substrates.

Organic cation transporter (OCT) 2 mediates the entry step for organic cation secretion by renal proximal tubule cells and is a site of unwanted drug-drug interactions (DDIs). But reliance on decision tree-based predictions of DDIs at OCT2 that depend on IC50 values can be suspect because they can be influenced by choice of transported substrate; for example, IC50 values for the inhibition of metformin versus MPP transport can vary by 5- to 10-fold. However, it is not clear whether the substrate dependence of a ligand interaction is common among OCT2 substrates. To address this question, we screened the inhibitory effectiveness of 20 µM concentrations of several hundred compounds against OCT2-mediated uptake of six structurally distinct substrates: MPP, metformin, N,N,N-trimethyl-2-[methyl(7-nitrobenzo[c][1,2,5]oxadiazol-4-yl)amino]ethanaminium (NBD-MTMA), TEA, cimetidine, and 4-4-dimethylaminostyryl-N-methylpyridinium (ASP). Of these, MPP transport was least sensitive to inhibition. IC50 values for 20 structurally diverse compounds confirmed this profile, with IC50 values for MPP averaging 6-fold larger than those for the other substrates. Bayesian machine-learning models of ligand-induced inhibition displayed generally good statistics after cross-validation and external testing. Applying our ASP model to a previously published large-scale screening study for inhibition of OCT2-mediated ASP transport resulted in comparable statistics, with approximately 75% of "active" inhibitors predicted correctly. The differential sensitivity of MPP transport to inhibition suggests that multiple ligands can interact simultaneously with OCT2 and supports the recommendation that MPP not be used as a test substrate for OCT2 screening. Instead, metformin appears to be a comparatively representative OCT2 substrate for both in vitro and in vivo (clinical) use.

Data Mining and Computational Modeling of High-Throughput Screening Datasets.

We are now seeing the benefit of investments made over the last decade in high-throughput screening (HTS) that is resulting in large structure activity datasets entering public and open databases such as ChEMBL and PubChem. The growth of academic HTS screening centers and the increasing move to academia for early stage drug discovery suggests a great need for the informatics tools and methods to mine such data and learn from it. Collaborative Drug Discovery, Inc. (CDD) has developed a number of tools for storing, mining, securely and selectively sharing, as well as learning from such HTS data. We present a new web based data mining and visualization module directly within the CDD Vault platform for high-throughput drug discovery data that makes use of a novel technology stack following modern reactive design principles. We also describe CDD Models within the CDD Vault platform that enables researchers to share models, share predictions from models, and create models from distributed, heterogeneous data. Our system is built on top of the Collaborative Drug Discovery Vault Activity and Registration data repository ecosystem which allows users to manipulate and visualize thousands of molecules in real time. This can be performed in any browser on any platform. In this chapter we present examples of its use with public datasets in CDD Vault. Such approaches can complement other cheminformatics tools, whether open source or commercial, in providing approaches for data mining and modeling of HTS data.

Comparing and Validating Machine Learning Models for Mycobacterium tuberculosis Drug Discovery.

Tuberculosis is a global health dilemma. In 2016, the WHO reported 10.4 million incidences and 1.7 million deaths. The need to develop new treatments for those infected with Mycobacterium tuberculosis ( Mtb) has led to many large-scale phenotypic screens and many thousands of new active compounds identified in vitro. However, with limited funding, efforts to discover new active molecules against Mtb needs to be more efficient. Several computational machine learning approaches have been shown to have good enrichment and hit rates. We have curated small molecule Mtb data and developed new models with a total of 18,886 molecules with activity cutoffs of 10 µM, 1 µM, and 100 nM. These data sets were used to evaluate different machine learning methods (including deep learning) and metrics and to generate predictions for additional molecules published in 2017. One Mtb model, a combined in vitro and in vivo data Bayesian model at a 100 nM activity yielded the following metrics for 5-fold cross validation: accuracy = 0.88, precision = 0.22, recall = 0.91, specificity = 0.88, kappa = 0.31, and MCC = 0.41. We have also curated an evaluation set ( n = 153 compounds) published in 2017, and when used to test our model, it showed the comparable statistics (accuracy = 0.83, precision = 0.27, recall = 1.00, specificity = 0.81, kappa = 0.36, and MCC = 0.47). We have also compared these models with additional machine learning algorithms showing Bayesian machine learning models constructed with literature Mtb data generated by different laboratories generally were equivalent to or outperformed deep neural networks with external test sets. Finally, we have also compared our training and test sets to show they were suitably diverse and different in order to represent useful evaluation sets. Such Mtb machine learning models could help prioritize compounds for testing in vitro and in vivo.

Understanding the patient experience of health care-associated infection: A qualitative systematic review.

BACKGROUND: The global burden of health care-associated infection (HAI) is well recognized; what is less well known is the impact HAI has on patients. To develop acceptable, effective interventions, greater understanding of patients' experience of HAI is needed. This qualitative systematic review sought to explore adult patients' experiences of common HAIs. METHODS: Five databases were searched. Search terms were combined for qualitative research, HAI terms, and patient experience. Study selection was conducted by 2 researchers using prespecified criteria. Critical Appraisal Skills Programme quality appraisal tools were used. Internationally recognized Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were applied. The Noblit and Hare (1988) approach to meta-synthesis was adopted. RESULTS: Seventeen studies (2001-2017) from 5 countries addressing 5 common types of HAI met the inclusion criteria. Four interrelated themes emerged: the continuum of physical and emotional responses, experiencing the response of health care professionals, adapting to life with an HAI, and the complex cultural context of HAI. CONCLUSIONS: The impact of different HAIs may vary; however, there are many similarities in the experience recounted by patients. The biosociocultural context of contagion was graphically expressed, with potential impact on social relationships and professional interactions highlighted. Further research to investigate contemporary patient experience in an era of antimicrobial resistance is warranted.

Collaborative drug discovery for More Medicines for Tuberculosis (MM4TB).

Neglected disease drug discovery is generally poorly funded compared with major diseases and hence there is an increasing focus on collaboration and precompetitive efforts such as public-private partnerships (PPPs). The More Medicines for Tuberculosis (MM4TB) project is one such collaboration funded by the EU with the goal of discovering new drugs for tuberculosis. Collaborative Drug Discovery has provided a commercial web-based platform called CDD Vault which is a hosted collaborative solution for securely sharing diverse chemistry and biology data. Using CDD Vault alongside other commercial and free cheminformatics tools has enabled support of this and other large collaborative projects, aiding drug discovery efforts and fostering collaboration. We will describe CDD's efforts in assisting with the MM4TB project.

Machine Learning Model Analysis and Data Visualization with Small Molecules Tested in a Mouse Model of Mycobacterium tuberculosis Infection (2014-2015).

The renewed urgency to develop new treatments for Mycobacterium tuberculosis (Mtb) infection has resulted in large-scale phenotypic screening and thousands of new active compounds in vitro. The next challenge is to identify candidates to pursue in a mouse in vivo efficacy model as a step to predicting clinical efficacy. We previously analyzed over 70 years of this mouse in vivo efficacy data, which we used to generate and validate machine learning models. Curation of 60 additional small molecules with in vivo data published in 2014 and 2015 was undertaken to further test these models. This represents a much larger test set than for the previous models. Several computational approaches have now been applied to analyze these molecules and compare their molecular properties beyond those attempted previously. Our previous machine learning models have been updated, and a novel aspect has been added in the form of mouse liver microsomal half-life (MLM t1/2) and in vitro-based Mtb models incorporating cytotoxicity data that were used to predict in vivo activity for comparison. Our best Mtb in vivo models possess fivefold ROC values > 0.7, sensitivity > 80%, and concordance > 60%, while the best specificity value is >40%. Use of an MLM t1/2 Bayesian model affords comparable results for scoring the 60 compounds tested. Combining MLM stability and in vitro Mtb models in a novel consensus workflow in the best cases has a positive predicted value (hit rate) > 77%. Our results indicate that Bayesian models constructed with literature in vivo Mtb data generated by different laboratories in various mouse models can have predictive value and may be used alongside MLM t1/2 and in vitro-based Mtb models to assist in selecting antitubercular compounds with desirable in vivo efficacy. We demonstrate for the first time that consensus models of any kind can be used to predict in vivo activity for Mtb. In addition, we describe a new clustering method for data visualization and apply this to the in vivo training and test data, ultimately making the method accessible in a mobile app.

Open Source Bayesian Models. 3. Composite Models for Prediction of Binned Responses.

Bayesian models constructed from structure-derived fingerprints have been a popular and useful method for drug discovery research when applied to bioactivity measurements that can be effectively classified as active or inactive. The results can be used to rank candidate structures according to their probability of activity, and this ranking benefits from the high degree of interpretability when structure-based fingerprints are used, making the results chemically intuitive. Besides selecting an activity threshold, building a Bayesian model is fast and requires few or no parameters or user intervention. The method also does not suffer from such acute overtraining problems as quantitative structure-activity relationships or quantitative structure-property relationships (QSAR/QSPR). This makes it an approach highly suitable for automated workflows that are independent of user expertise or prior knowledge of the training data. We now describe a new method for creating a composite group of Bayesian models to extend the method to work with multiple states, rather than just binary. Incoming activities are divided into bins, each covering a mutually exclusive range of activities. For each of these bins, a Bayesian model is created to model whether or not the compound belongs in the bin. Analyzing putative molecules using the composite model involves making a prediction for each bin and examining the relative likelihood for each assignment, for example, highest value wins. The method has been evaluated on a collection of hundreds of data sets extracted from ChEMBL v20 and validated data sets for ADME/Tox and bioactivity.

Making Transporter Models for Drug-Drug Interaction Prediction Mobile.

The past decade has seen increased numbers of studies publishing ligand-based computational models for drug transporters. Although they generally use small experimental data sets, these models can provide insights into structure-activity relationships for the transporter. In addition, such models have helped to identify new compounds as substrates or inhibitors of transporters of interest. We recently proposed that many transporters are promiscuous and may require profiling of new chemical entities against multiple substrates for a specific transporter. Furthermore, it should be noted that virtually all of the published ligand-based transporter models are only accessible to those involved in creating them and, consequently, are rarely shared effectively. One way to surmount this is to make models shareable or more accessible. The development of mobile apps that can access such models is highlighted here. These apps can be used to predict ligand interactions with transporters using Bayesian algorithms. We used recently published transporter data sets (MATE1, MATE2K, OCT2, OCTN2, ASBT, and NTCP) to build preliminary models in a commercial tool and in open software that can deliver the model in a mobile app. In addition, several transporter data sets extracted from the ChEMBL database were used to illustrate how such public data and models can be shared. Predicting drug-drug interactions for various transporters using computational models is potentially within reach of anyone with an iPhone or iPad. Such tools could help prioritize which substrates should be used for in vivo drug-drug interaction testing and enable open sharing of models.