Search

Abstract Conformal prediction has seen many applications in pharmaceutical science, being able to calibrate outputs of machine learning models and producing valid prediction intervals. We here present the open source software CPSign that is a complete implementation of conformal prediction for cheminformatics modeling. CPSign implements inductive and transductive conformal prediction for classification and regression, and probabilistic prediction with the Venn-ABERS methodology. The main chemical representation is signatures but other types of descriptors are also supported. The main modeling methodology is support vector machines (SVMs), but additional modeling methods are supported via an extension mechanism, e.g. DeepLearning4J models. We also describe features for visualizing results from conformal models including calibration and efficiency plots, as well as features to publish predictive models as REST services. We compare CPSign against other common cheminformatics modeling approaches including random forest, and a directed message-passing neural network. The results show that CPSign produces robust predictive performance with comparative predictive efficiency, with superior runtime and lower hardware requirements compared to neural network based models. CPSign has been used in several studies and is in production-use in multiple organizations. The ability to work directly with chemical input files, perform descriptor calculation and modeling with SVM in the conformal prediction framework, with a single software package having a low footprint and fast execution time makes CPSign a convenient and yet flexible package for training, deploying, and predicting on chemical data. CPSign can be downloaded from GitHub at https://github.com/arosbio/cpsign . Scientific contribution CPSign provides a single software that allows users to perform data preprocessing, modeling and make predictions directly on chemical structures, using conformal and probabilistic prediction. Building and evaluating new models can be achieved at a high abstraction level, without sacrificing flexibility and predictive performance—showcased with a method evaluation against contemporary modeling approaches, where CPSign performs on par with a state-of-the-art deep learning based model.

Having access to sufficient data is essential in order to train accurate machine learning models, but much data is not publicly available. In drug discovery this is particularly evident, as much data is withheld at pharmaceutical companies for various reasons. Federated Learning (FL) aims at training a joint model between multiple parties but without disclosing data between the parties. In this work, we leverage Federated Learning to predict compound Mechanism of Action (MoA) using fluorescence image data from cell painting. Our study evaluates the effectiveness and efficiency of FL, comparing to non-collaborative and data-sharing collaborative learning in diverse scenarios. Specifically, we investigate the impact of data heterogeneity across participants on MoA prediction, an essential concern in real-life applications of FL, and demonstrate the benefits for all involved parties. This work highlights the potential of federated learning in multi-institutional collaborative machine learning for drug discovery and assessment of chemicals, offering a promising avenue to overcome data-sharing constraints.

In the European regulatory context, rodent in vivo studies are the predominant source of neurotoxicity information. Although they form a cornerstone of neurotoxicological assessments, they are costly and the topic of ethical debate. While the public expects chemicals and products to be safe for the developing and mature nervous systems, considerable numbers of chemicals in commerce have not, or only to a limited extent, been assessed for their potential to cause neurotoxicity. As such, there is a societal push toward the replacement of animal models with in vitro or alternative methods. New approach methods (NAMs) can contribute to the regulatory knowledge base, increase chemical safety, and modernize chemical hazard and risk assessment. Provided they reach an acceptable level of regulatory relevance and reliability, NAMs may be considered as replacements for specific in vivo studies. The European Partnership for the Assessment of Risks from Chemicals (PARC) addresses challenges to the development and implementation of NAMs in chemical risk assessment. In collaboration with regulatory agencies, Project 5.2.1e (Neurotoxicity) aims to develop and evaluate NAMs for developmental neurotoxicity (DNT) and adult neurotoxicity (ANT) and to understand the applicability domain of specific NAMs for the detection of endocrine disruption and epigenetic perturbation. To speed up assay time and reduce costs, we identify early indicators of later-onset effects. Ultimately, we will assemble second-generation developmental neurotoxicity and first-generation adult neurotoxicity test batteries, both of which aim to provide regulatory hazard and risk assessors and industry stakeholders with robust, speedy, lower-cost, and informative next-generation hazard and risk assessment tools.

Abstract The applicability domain of machine learning models trained on structural fingerprints for the prediction of biological endpoints is often limited by the lack of diversity of chemical space of the training data. In this work, we developed similarity-based merger models which combined the outputs of individual models trained on cell morphology (based on Cell Painting) and chemical structure (based on chemical fingerprints) and the structural and morphological similarities of the compounds in the test dataset to compounds in the training dataset. We applied these similarity-based merger models using logistic regression models on the predictions and similarities as features and predicted assay hit calls of 177 assays from ChEMBL, PubChem and the Broad Institute (where the required Cell Painting annotations were available). We found that the similarity-based merger models outperformed other models with an additional 20% assays (79 out of 177 assays) with an AUC > 0.70 compared with 65 out of 177 assays using structural models and 50 out of 177 assays using Cell Painting models. Our results demonstrated that similarity-based merger models combining structure and cell morphology models can more accurately predict a wide range of biological assay outcomes and further expanded the applicability domain by better extrapolating to new structural and morphology spaces. Graphical Abstract

Microplates are indispensable in large-scale biomedical experiments but the physical location of samples and controls on the microplate can significantly affect the resulting data and quality metric values. We introduce a new method based on constraint programming for designing microplate layouts that reduces unwanted bias and limits the impact of batch effects after error correction and normalisation. We demonstrate that our method applied to dose-response experiments leads to more accurate regression curves and lower errors when estimating IC50/EC50, and for drug screening leads to increased precision, when compared to random layouts. It also reduces the risk of inflated scores from common microplate quality assessment metrics such as Z′ factor and SSMD. We make our method available via a suite of tools (PLAID) including a reference constraint model, a web application, and Python notebooks to evaluate and compare designs when planning microplate experiments.

The mechanism of action (MoA) of a compound describes the biological interaction through which it produces a pharmacological effect. Multiple data sources can be used for the purpose of predicting MoA, including compound structural information, and various assays, such as those based on cell morphology, transcriptomics and metabolomics. In the present study we explored the benefits and potential additive/synergistic effects of combining structural information, in the form of Morgan fingerprints, and morphological information, in the form of five-channel Cell Painting image data. For a set of 10 well represented MoA classes, we compared the performance of deep learning models trained on the two datasets separately versus a model trained on both datasets simultaneously. On a held-out test set we obtained a macro-averaged F1 score of 0.58 when training on only the structural data, 0.81 when training on only the image data, and 0.92 when training on both together. Thus indicating clear additive/synergistic effects and highlighting the benefit of integrating multiple data sources for MoA prediction.

Toxicology has been an active research field for many decades, with academic, industrial and government involvement. Modern omics and computational approaches are changing the field, from merely disease-specific observational models into target-specific predictive models. Traditionally, toxicology has strong links with other fields such as biology, chemistry, pharmacology, and medicine. With the rise of synthetic and new engineered materials, alongside ongoing prioritisation needs in chemical risk assessment for existing chemicals, early predictive evaluations are becoming of utmost importance to both scientific and regulatory purposes. ELIXIR is an intergovernmental organisation that brings together life science resources from across Europe. To coordinate the linkage of various life science efforts around modern predictive toxicology, the establishment of a new ELIXIR Community is seen as instrumental. In the past few years, joint efforts, building on incidental overlap, have been piloted in the context of ELIXIR. For example, the EU-ToxRisk, diXa, HeCaToS, transQST, and the nanotoxicology community have worked with the ELIXIR TeSS, Bioschemas, and Compute Platforms and activities. In 2018, a core group of interested parties wrote a proposal, outlining a sketch of what this new ELIXIR Toxicology Community would look like. A recent workshop (held September 30th to October 1st, 2020) extended this into an ELIXIR Toxicology roadmap and a shortlist of limited investment-high gain collaborations to give body to this new community. This Whitepaper outlines the results of these efforts and defines our vision of the ELIXIR Toxicology Community and how it complements other ELIXIR activities.

Artificial intelligence prediction accuracy can be reduced with new data. Here, the authors utilise conformal prediction to reduce incorrect predictions in histopathological analysis of prostate cancer biopsies.

The biomedical research community is investing heavily in biomedical cloud platforms. Cloud computing holds great promise for addressing challenges with big data and ensuring reproducibility in biology. However, despite their advantages, cloud platforms in and of themselves do not automatically support FAIRness. The global push to develop biomedical cloud platforms has led to new challenges, including platform lock-in, difficulty integrating across platforms, and duplicated effort for both users and developers. Here, we argue that these difficulties are systemic and emerge from incentives that encourage development effort on self-sufficient platforms and data repositories instead of interoperable microservices. We argue that many of these issues would be alleviated by prioritizing microservices and access to modular data in smaller chunks or summarized form. We propose that emphasizing modularity and interoperability would lead to a more powerful Unix-like ecosystem of web services for biomedical analysis and data retrieval. We challenge funders, developers, and researchers to support a vision to improve interoperability through microservices as the next generation of cloud-based bioinformatics.

Fluorescence staining techniques, such as Cell Painting, together with fluorescence microscopy have proven invaluable for visualizing and quantifying the effects that drugs and other perturbations have on cultured cells. However, fluorescence microscopy is expensive, time-consuming, labor-intensive, and the stains applied can be cytotoxic, interfering with the activity under study. The simplest form of microscopy, brightfield microscopy, lacks these downsides, but the images produced have low contrast and the cellular compartments are difficult to discern. Nevertheless, by harnessing deep learning, these brightfield images may still be sufficient for various predictive purposes. In this study, we compared the predictive performance of models trained on fluorescence images to those trained on brightfield images for predicting the mechanism of action (MoA) of different drugs. We also extracted CellProfiler features from the fluorescence images and used them to benchmark the performance. Overall, we found comparable and largely correlated predictive performance for the two imaging modalities. This is promising for future studies of MoAs in time-lapse experiments for which using fluorescence images is problematic. Explorations based on explainable AI techniques also provided valuable insights regarding compounds that were better predicted by one modality over the other.

In past times, the analysis of endocrine disrupting properties of chemicals has mainly been focused on (anti-)estrogenic or (anti-)androgenic properties, as well as on aspects of steroidogenesis and the modulation of thyroid signaling. More recently, disruption of energy metabolism and related signaling pathways by exogenous substances, so-called metabolism-disrupting chemicals (MDCs) have come into focus. While general effects such as body and organ weight changes are routinely monitored in animal studies, there is a clear lack of mechanistic test systems to determine and characterize the metabolism-disrupting potential of chemicals. In order to contribute to filling this gap, one of the project within EU-funded Partnership for the Assessment of Risks of Chemicals (PARC) aims at developing novel in vitro methods for the detection of endocrine metabolic disruptors. Efforts will comprise projects related to specific signaling pathways, for example, involving mTOR or xenobiotic-sensing nuclear receptors, studies on hepatocytes, adipocytes and pancreatic beta cells covering metabolic and morphological endpoints, as well as metabolism-related zebrafish-based tests as an alternative to classic rodent bioassays. This paper provides an overview of the approaches and methods of these PARC projects and how this will contribute to the improvement of the toxicological toolbox to identify substances with endocrine disrupting properties and to decipher their mechanisms of action.

Summary: Progressive multiple sclerosis (PMS) is currently diagnosed retrospectively. Here, we work toward a set of biomarkers that could assist in early diagnosis of PMS. A selection of cerebrospinal fluid metabolites (n = 15) was shown to differentiate between PMS and its preceding phenotype in an independent cohort (AUC = 0.93). Complementing the classifier with conformal prediction showed that highly confident predictions could be made, and that three out of eight patients developing PMS within three years of sample collection were predicted as PMS at that time point. Finally, this methodology was applied to PMS patients as part of a clinical trial for intrathecal treatment with rituximab. The methodology showed that 68% of the patients decreased their similarity to the PMS phenotype one year after treatment. In conclusion, the inclusion of confidence predictors contributes with more information compared to traditional machine learning, and this information is relevant for disease monitoring.

Abstract Determining the levels of protein–protein interactions is essential for the analysis of signaling within the cell, characterization of mutation effects, protein function and activation in health and disease, among others. Herein, we describe MolBoolean – a method to detect interactions between endogenous proteins in various subcellular compartments, utilizing antibody-DNA conjugates for identification and signal amplification. In contrast to proximity ligation assays, MolBoolean simultaneously indicates the relative abundances of protein A and B not interacting with each other, as well as the pool of A and B proteins that are proximal enough to be considered an AB complex. MolBoolean is applicable both in fixed cells and tissue sections. The specific and quantifiable data that the method generates provide opportunities for both diagnostic use and medical research.

Cell Painting, gene expression, and chemical structural data are used to examine the differences between mitochondrial toxicants and non-toxicants and enhance the detection of mitotoxic compounds for future drug discovery.

Abstract Polypharmacy refers to the administration of multiple drugs on a daily basis. It has demonstrated effectiveness in treating many complex diseases , but it has a higher risk of adverse drug reactions. Hence, the prediction of polypharmacy side effects is an essential step in drug testing, especially for new drugs. This paper shows that the current knowledge graph (KG) based state-of-the-art approach to polypharmacy side effect prediction does not work well for new drugs, as they have a low number of known connections in the KG. We propose a new method , SimVec, that solves this problem by enhancing the KG structure with a structure-aware node initialization and weighted drug similarity edges. We also devise a new 3-step learning process, which iteratively updates node embeddings related to side effects edges, similarity edges, and drugs with limited knowledge. Our model significantly outperforms existing KG-based models. Additionally, we examine the problem of negative relations generation and show that the cache-based approach works best for polypharmacy tasks.

Abstract Comparing chemical structures to infer protein targets and functions is a common approach, but basing comparisons on chemical similarity alone can be misleading. Here we present a methodology for predicting target protein clusters using deep neural networks. The model is trained on clusters of compounds based on similarities calculated from combined compound-protein and protein-protein interaction data using a network topology approach. We compare several deep learning architectures including both convolutional and recurrent neural networks. The best performing method, the recurrent neural network architecture MolPMoFiT, achieved an F1 score approaching 0.9 on a held-out test set of 8907 compounds. In addition, in-depth analysis on a set of eleven well-studied chemical compounds with known functions showed that predictions were justifiable for all but one of the chemicals. Four of the compounds, similar in their molecular structure but with dissimilarities in their function, revealed advantages of our method compared to using chemical similarity.

Abstract Confidence predictors can deliver predictions with the associated confidence required for decision making and can play an important role in drug discovery and toxicity predictions. In this work we investigate a recently introduced version of conformal prediction, synergy conformal prediction, focusing on the predictive performance when applied to bioactivity data. We compare the performance to other variants of conformal predictors for multiple partitioned datasets and demonstrate the utility of synergy conformal predictors for federated learning where data cannot be pooled in one location. Our results show that synergy conformal predictors based on training data randomly sampled with replacement can compete with other conformal setups, while using completely separate training sets often results in worse performance. However, in a federated setup where no method has access to all the data, synergy conformal prediction is shown to give promising results. Based on our study, we conclude that synergy conformal predictors are a valuable addition to the conformal prediction toolbox.

Abstract Background The emergence and continued global spread of the current COVID-19 pandemic has highlighted the need for methods to identify novel or repurposed therapeutic drugs in a fast and effective way. Despite the availability of methods for the discovery of antiviral drugs, the majority tend to focus on the effects of such drugs on a given virus, its constituent proteins, or enzymatic activity, often neglecting the consequences on host cells. This may lead to partial assessment of the efficacy of the tested anti-viral compounds, as potential toxicity impacting the overall physiology of host cells may mask the effects of both viral infection and drug candidates. Here we present a method able to assess the general health of host cells based on morphological profiling, for untargeted phenotypic drug screening against viral infections. Results We combine Cell Painting with antibody-based detection of viral infection in a single assay. We designed an image analysis pipeline for segmentation and classification of virus-infected and non-infected cells, followed by extraction of morphological properties. We show that this methodology can successfully capture virus-induced phenotypic signatures of MRC-5 human lung fibroblasts infected with human coronavirus 229E (CoV-229E). Moreover, we demonstrate that our method can be used in phenotypic drug screening using a panel of nine host- and virus-targeting antivirals. Treatment with effective antiviral compounds reversed the morphological profile of the host cells towards a non-infected state. Conclusions The phenomics approach presented here, which makes use of a modified Cell Painting protocol by incorporating an anti-virus antibody stain, can be used for the unbiased morphological profiling of virus infection on host cells. The method can identify antiviral reference compounds, as well as novel antivirals, demonstrating its suitability to be implemented as a strategy for antiviral drug repurposing and drug discovery.

Objective: The aim of this project was to implement long-read sequencing for BCR-ABL1 TKI resistance mutation screening in a clinical setting for patients undergoing treatment for chronic myeloid leukemia. Materials and Methods: Processes were established for registering and transferring samples from the clinic to an academic sequencing facility for long-read sequencing. An automated analysis pipeline for detecting mutations was established, and an information system was implemented comprising features for data management, analysis and visualization. Clinical validation was performed by identifying BCR-ABL1 TKI resistance mutations by Sanger and long-read sequencing in parallel. The developed software is available as open source via GitHub at https://github.com/pharmbio/clamp Results: The information system enabled traceable transfer of samples from the clinic to the sequencing facility, robust and automated analysis of the long-read sequence data, and communication of results from sequence analysis in a reporting format that could be easily interpreted and acted upon by clinical experts. In a validation study, all 17 resistance mutations found by Sanger sequencing were also detected by long-read sequencing. An additional 16 mutations were found only by long-read sequencing, all of them with frequencies below the limit of detection for Sanger sequencing. The clonal distributions of co-existing mutations were automatically resolved through the long-read data analysis. After the implementation and validation, the clinical laboratory switched their routine protocol from using Sanger to long-read sequencing for this application. Conclusions: Long-read sequencing delivers results with higher sensitivity compared to Sanger sequencing and enables earlier detection of emerging TKI resistance mutations. The developed processes, analysis workflow, and software components lower barriers for adoption and could be extended to other applications.

Abstract Machine learning methods are widely used in drug discovery and toxicity prediction. While showing overall good performance in cross-validation studies, their predictive power (often) drops in cases where the query samples have drifted from the training data’s descriptor space. Thus, the assumption for applying machine learning algorithms, that training and test data stem from the same distribution, might not always be fulfilled. In this work, conformal prediction is used to assess the calibration of the models. Deviations from the expected error may indicate that training and test data originate from different distributions. Exemplified on the Tox21 datasets, composed of chronologically released Tox21Train, Tox21Test and Tox21Score subsets, we observed that while internally valid models could be trained using cross-validation on Tox21Train, predictions on the external Tox21Score data resulted in higher error rates than expected. To improve the prediction on the external sets, a strategy exchanging the calibration set with more recent data, such as Tox21Test, has successfully been introduced. We conclude that conformal prediction can be used to diagnose data drifts and other issues related to model calibration. The proposed improvement strategy—exchanging the calibration data only—is convenient as it does not require retraining of the underlying model.

Abstract Background Identifying and assessing ligand-target binding is a core component in early drug discovery as one or more unwanted interactions may be associated with safety issues. Contributions We present an open-source, extendable web service for predicting target profiles with confidence using machine learning for a panel of 7 targets, where models are trained on molecular docking scores from a large virtual library. The method uses conformal prediction to produce valid measures of prediction efficiency for a particular confidence level. The service also offers the possibility to dock chemical structures to the panel of targets with QuickVina on individual compound basis. Results The docking procedure and resulting models were validated by docking well-known inhibitors for each of the 7 targets using QuickVina. The model predictions showed comparable performance to molecular docking scores against an external validation set. The implementation as publicly available microservices on Kubernetes ensures resilience, scalability, and extensibility.

Abstract The reproducibility of experiments has been a long standing impediment for further scientific progress. Computational methods have been instrumental in drug discovery efforts owing to its multifaceted utilization for data collection, pre-processing, analysis and inference. This article provides an in-depth coverage on the reproducibility of computational drug discovery. This review explores the following topics: (1) the current state-of-the-art on reproducible research, (2) research documentation (e.g. electronic laboratory notebook, Jupyter notebook, etc.), (3) science of reproducible research (i.e. comparison and contrast with related concepts as replicability, reusability and reliability), (4) model development in computational drug discovery, (5) computational issues on model development and deployment, (6) use case scenarios for streamlining the computational drug discovery protocol. In computational disciplines, it has become common practice to share data and programming codes used for numerical calculations as to not only facilitate reproducibility, but also to foster collaborations (i.e. to drive the project further by introducing new ideas, growing the data, augmenting the code, etc.). It is therefore inevitable that the field of computational drug design would adopt an open approach towards the collection, curation and sharing of data/code.

Abstract Ligand-based predictive modeling is widely used to generate predictive models aiding decision making in e.g. drug discovery projects. With growing data sets and requirements on low modeling time comes the necessity to analyze data sets efficiently to support rapid and robust modeling. In this study we analyzed four data sets and studied the efficiency of machine learning methods on sparse data structures, utilizing Morgan fingerprints of different radii and hash sizes, and compared with molecular signatures descriptor of different height. We specifically evaluated the effect these parameters had on modeling time, predictive performance, and memory requirements using two implementations of random forest; Scikit-learn as well as FEST. We also compared with a support vector machine implementation. Our results showed that unhashed fingerprints yield significantly better accuracy than hashed fingerprints ($$p \le 0.05$$ p≤0.05 ), with no pronounced deterioration in modeling time and memory usage. Furthermore, the fast execution and low memory usage of the FEST algorithm suggest that it is a good alternative for large, high dimensional sparse data. Both support vector machines and random forest performed equally well but results indicate that the support vector machine was better at using the extra information from larger values of the Morgan fingerprint’s radius.

Abstract Lipophilicity is a major determinant of ADMET properties and overall suitability of drug candidates. We have developed large-scale models to predict water–octanol distribution coefficient (logD) for chemical compounds, aiding drug discovery projects. Using ACD/logD data for 1.6 million compounds from the ChEMBL database, models are created and evaluated by a support-vector machine with a linear kernel using conformal prediction methodology, outputting prediction intervals at a specified confidence level. The resulting model shows a predictive ability of $$\hbox {Q}^{2}=0.973$$ Q2=0.973 and with the best performing nonconformity measure having median prediction interval of $$\pm ~0.39$$ ±0.39 log units at 80% confidence and $$\pm ~0.60$$ ±0.60 log units at 90% confidence. The model is available as an online service via an OpenAPI interface, a web page with a molecular editor, and we also publish predictive values at 90% confidence level for 91 M PubChem structures in RDF format for download and as an URI resolver service.

Abstract Background Docking and scoring large libraries of ligands against target proteins forms the basis of structure-based virtual screening. The problem is trivially parallelizable, and calculations are generally carried out on computer clusters or on large workstations in a brute force manner, by docking and scoring all available ligands. Contribution In this study we propose a strategy that is based on iteratively docking a set of ligands to form a training set, training a ligand-based model on this set, and predicting the remainder of the ligands to exclude those predicted as ‘low-scoring’ ligands. Then, another set of ligands are docked, the model is retrained and the process is repeated until a certain model efficiency level is reached. Thereafter, the remaining ligands are docked or excluded based on this model. We use SVM and conformal prediction to deliver valid prediction intervals for ranking the predicted ligands, and Apache Spark to parallelize both the docking and the modeling. Results We show on 4 different targets that conformal prediction based virtual screening (CPVS) is able to reduce the number of docked molecules by 62.61% while retaining an accuracy for the top 30 hits of 94% on average and a speedup of 3.7. The implementation is available as open source via GitHub (https://github.com/laeeq80/spark-cpvs) and can be run on high-performance computers as well as on cloud resources.

Bisphenols are important environmental pollutants that are extensively studied due to different detrimental effects, while the molecular mechanisms behind these effects are less well understood. Like other environmental pollutants, bisphenols are being tested in various experimental models, creating large expression datasets found in open access storage. The meta-analysis of such datasets is, however, very complicated for various reasons. Here, we developed an integrating statistical and machine-learning model approach for the meta-analysis of bisphenol A (BPA) exposure datasets from different mouse tissues. We constructed three joint datasets following three different strategies for dataset integration: in particular, using all common genes from the datasets, uncorrelated, and not co-expressed genes, respectively. By applying machine learning methods to these datasets, we identified genes whose expression was significantly affected in all of the BPA microanalysis data tested; those involved in the regulation of cell survival include: Tnfr2, Hgf-Met, Agtr1a, Bdkrb2; signaling through Mapk8 (Jnk1)); DNA repair (Hgf-Met, Mgmt); apoptosis (Tmbim6, Bcl2, Apaf1); and cellular junctions (F11r, Cldnd1, Ctnd1 and Yes1). Our results highlight the benefit of combining existing datasets for the integrated analysis of a specific topic when individual datasets are limited in size.

Abstract Background Biological sciences are characterised not only by an increasing amount but also the extreme complexity of its data. This stresses the need for efficient ways of integrating these data in a coherent description of biological systems. In many cases, biological data needs organization before integration. This is not seldom a collaborative effort, and it is thus important that tools for data integration support a collaborative way of working. Wiki systems with support for structured semantic data authoring, such as Semantic MediaWiki, provide a powerful solution for collaborative editing of data combined with machine-readability, so that data can be handled in an automated fashion in any downstream analyses. Semantic MediaWiki lacks a built-in data import function though, which hinders efficient round-tripping of data between interoperable Semantic Web formats such as RDF and the internal wiki format. Results To solve this deficiency, the RDFIO suite of tools is presented, which supports importing of RDF data into Semantic MediaWiki, with metadata needed to export it again in the same RDF format, or ontology. Additionally, the new functionality enables mash-ups of automated data imports combined with manually created data presentations. The application of the suite of tools is demonstrated by importing drug discovery related data about rare diseases from Orphanet and acid dissociation constants from Wikidata. The RDFIO suite of tools is freely available for download via pharmb.io/project/rdfio . Conclusions Through a set of biomedical demonstrators, it is demonstrated how the new functionality enables a number of usage scenarios where the interoperability of SMW and the wider Semantic Web is leveraged for biomedical data sets, to create an easy to use and flexible platform for exploring and working with biomedical data.

Abstract Background The Chemistry Development Kit (CDK) is a widely used open source cheminformatics toolkit, providing data structures to represent chemical concepts along with methods to manipulate such structures and perform computations on them. The library implements a wide variety of cheminformatics algorithms ranging from chemical structure canonicalization to molecular descriptor calculations and pharmacophore perception. It is used in drug discovery, metabolomics, and toxicology. Over the last 10 years, the code base has grown significantly, however, resulting in many complex interdependencies among components and poor performance of many algorithms. Results We report improvements to the CDK v2.0 since the v1.2 release series, specifically addressing the increased functional complexity and poor performance. We first summarize the addition of new functionality, such atom typing and molecular formula handling, and improvement to existing functionality that has led to significantly better performance for substructure searching, molecular fingerprints, and rendering of molecules. Second, we outline how the CDK has evolved with respect to quality control and the approaches we have adopted to ensure stability, including a code review mechanism. Conclusions This paper highlights our continued efforts to provide a community driven, open source cheminformatics library, and shows that such collaborative projects can thrive over extended periods of time, resulting in a high-quality and performant library. By taking advantage of community support and contributions, we show that an open source cheminformatics project can act as a peer reviewed publishing platform for scientific computing software. Graphical abstract CDK 2.0 provides new features and improved performance

The computational demands for scientific applications are continuously increasing. The emergence of cloud computing has enabled on-demand resource allocation. However, relying solely on infrastructure as a service does not achieve the degree of flexibility required by the scientific community. Here we present a microservice-oriented methodology, where scientific applications run in a distributed orchestration platform as software containers, referred to as on-demand, virtual research environments. The methodology is vendor agnostic and we provide an open source implementation that supports the major cloud providers, offering scalable management of scientific pipelines. We demonstrate applicability and scalability of our methodology in life science applications, but the methodology is general and can be applied to other scientific domains.

Oral bioavailability (F) is an essential determinant for the systemic exposure and dosing regimens of drug candidates. F is determined by numerous processes, and computational predictions of human estimates have so far shown limited results. We describe a new methodology where F in humans is predicted directly from chemical structure using an integrated strategy combining 9 machine learning models, 3 sets of structural alerts, and 2 physiologically-based pharmacokinetic models. We evaluate the model on a benchmark dataset consisting of 184 compounds, obtaining a predictive accuracy (Q2) of 0.50, which is successful according to a pharmaceutical industry proposal. Twenty-seven compounds were found (beforehand) to be outside the main applicability domain for the model. We compare our results with interspecies correlations (rat, mouse and dog vs. human) using the same dataset, where animal vs. human-correlations (R2) were found to be 0.21 to 0.40 and maximum prediction errors were smaller than maximum interspecies differences. We conclude that our method has sufficient predictive accuracy to be practically useful with applications in human exposure and dose predictions, compound optimization and decision making, with potential to rationalize drug discovery and development and decrease failures and overexposures in early clinical trials with candidate drugs.

Ligand-based models can be used in drug discovery to obtain an early indication of potential off-target interactions that could be linked to adverse effects. Another application is to combine such models into a panel, allowing to compare and search for compounds with similar profiles. Most contemporary methods and implementations however lack valid measures of confidence in their predictions, and only provide point predictions. We here describe a methodology that uses Conformal Prediction for predicting off-target interactions, with models trained on data from 31 targets in the ExCAPE-DB dataset selected for their utility in broad early hazard assessment. Chemicals were represented by the signature molecular descriptor and support vector machines were used as the underlying machine learning method. By using conformal prediction, the results from predictions come in the form of confidence p-values for each class. The full pre-processing and model training process is openly available as scientific workflows on GitHub, rendering it fully reproducible. We illustrate the usefulness of the developed methodology on a set of compounds extracted from DrugBank. The resulting models are published online and are available via a graphical web interface and an OpenAPI interface for programmatic access.

Metabolomics, the youngest of the major omics technologies, is supported by an active community of researchers and infrastructure developers across Europe. To coordinate and focus efforts around infrastructure building for metabolomics within Europe, a workshop on the “Future of metabolomics in ELIXIR” was organised at Frankfurt Airport in Germany. This one-day strategic workshop involved representatives of ELIXIR Nodes, members of the PhenoMeNal consortium developing an e-infrastructure that supports workflow-based metabolomics analysis pipelines, and experts from the international metabolomics community. The workshop established metabolite identification as the critical area, where a maximal impact of computational metabolomics and data management on other fields could be achieved. In particular, the existing four ELIXIR Use Cases, where the metabolomics community - both industry and academia - would benefit most, and which could be exhaustively mapped onto the current five ELIXIR Platforms were discussed. This opinion article is a call for support for a new ELIXIR metabolomics Use Case, which aligns with and complements the existing and planned ELIXIR Platforms and Use Cases.

Metabolomics, the youngest of the major omics technologies, is supported by an active community of researchers and infrastructure developers across Europe. To coordinate and focus efforts around infrastructure building for metabolomics within Europe, a workshop on the “Future of metabolomics in ELIXIR” was organised at Frankfurt Airport in Germany. This one-day strategic workshop involved representatives of ELIXIR Nodes, members of the PhenoMeNal consortium developing an e-infrastructure that supports workflow-based metabolomics analysis pipelines, and experts from the international metabolomics community. The workshop established metabolite identification as the critical area, where a maximal impact of computational metabolomics and data management on other fields could be achieved. In particular, the existing four ELIXIR Use Cases, where the metabolomics community - both industry and academia - would benefit most, and which could be exhaustively mapped onto the current five ELIXIR Platforms were discussed. This opinion article is a call for support for a new ELIXIR metabolomics Use Case, which aligns with and complements the existing and planned ELIXIR Platforms and Use Cases.

To better understand the pathophysiological differences between secondary progressive multiple sclerosis (SPMS) and relapsing-remitting multiple sclerosis (RRMS), and to identify potential biomarkers of disease progression, we applied high-resolution mass spectrometry (HRMS) to investigate the metabolome of cerebrospinal fluid (CSF). The biochemical differences were determined using partial least squares discriminant analysis (PLS-DA) and connected to biochemical pathways as well as associated to clinical and radiological measures. Tryptophan metabolism was significantly altered, with perturbed levels of kynurenate, 5-hydroxytryptophan, 5-hydroxyindoleacetate, and N-acetylserotonin in SPMS patients compared with RRMS and controls. SPMS patients had altered kynurenine compared with RRMS patients, and altered indole-3-acetate compared with controls. Regarding the pyrimidine metabolism, SPMS patients had altered levels of uridine and deoxyuridine compared with RRMS and controls, and altered thymine and glutamine compared with RRMS patients. Metabolites from the pyrimidine metabolism were significantly associated with disability, disease activity and brain atrophy, making them of particular interest for understanding the disease mechanisms and as markers of disease progression. Overall, these findings are of importance for the characterization of the molecular pathogenesis of SPMS and support the hypothesis that the CSF metabolome may be used to explore changes that occur in the transition between the RRMS and SPMS pathologies.

Aromatase, the rate-limiting enzyme that catalyzes the conversion of androgen to estrogen, plays an essential role in the development of estrogen-dependent breast cancer. Side effects due to aromatase inhibitors (AIs) necessitate the pursuit of novel inhibitor candidates with high selectivity, lower toxicity and increased potency. Designing a novel therapeutic agent against aromatase could be achieved computationally by means of ligand-based and structure-based methods. For over a decade, we have utilized both approaches to design potential AIs for which quantitative structure–activity relationships and molecular docking were used to explore inhibitory mechanisms of AIs towards aromatase. However, such approaches do not consider the effects that aromatase variants have on different AIs. In this study, proteochemometrics modeling was applied to analyze the interaction space between AIs and aromatase variants as a function of their substructural and amino acid features. Good predictive performance was achieved, as rigorously verified by 10-fold cross-validation, external validation, leave-one-compound-out cross-validation, leave-one-protein-out cross-validation and Y-scrambling tests. The investigations presented herein provide important insights into the mechanisms of aromatase inhibitory activity that could aid in the design of novel potent AIs as breast cancer therapeutic agents.

In order to identify new ways to prevent, diagnose and treat diseases, biobanks systematically collect samples of human tissues and population-wide data on health and lifestyle. Efficient access to population biobank data and to biomaterial is crucial for development and marketing of new pharmaceutical products, especially in the area of personalised medicine. However, such access is hindered by legal and ethical constraints, and by the huge semantic diversity across different biobanks. To address these challenges, we have developed SAIL, a sophisticated metaportal for biobank data annotation across different collections and repositories, harmonised to allow cross-biobank searchability, while preserving the anonymity and privacy of the underlying data such that legal and ethical requirements are met. We describe the technological architecture and design of SAIL that allows us to meet these pressing challenges, and give an overview of the current functionality of the application. SAIL is available online at sail.simbioms.org, and it currently contains around 200 000 samples from 14 collections.

A unified proteochemometric (PCM) model for the prediction of the ability of drug-like chemicals to inhibit five major drug metabolizing CYP isoforms (i.e. CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A4) was created and made publicly available under the Bioclipse Decision Support open source system at www.cyp450model.org. In regards to the proteochemometric modeling we represented the chemical compounds by molecular signature descriptors and the CYP-isoforms by alignment-independent description of composition and transition of amino acid properties of their protein primary sequences. The entire training dataset contained 63 391 interactions and the best PCM model was obtained using signature descriptors of height 1, 2 and 3 and inducing the model with a support vector machine. The model showed excellent predictive ability with internal AUC = 0.923 and an external AUC = 0.940, as evaluated on a large external dataset. The advantage of PCM models is their extensibility making it possible to extend our model for new CYP isoforms and polymorphic CYP forms. A key benefit of PCM is that all proteins are confined in one single model, which makes it generally more stable and predictive as compared with single target models. The inclusion of the model in Bioclipse Decision Support makes it possible to make virtual instantaneous predictions (∼100 ms per prediction) while interactively drawing or modifying chemical structures in the Bioclipse chemical structure editor.

BACKGROUND: Reverse transcriptase is a major drug target in highly active antiretroviral therapy (HAART) against HIV, which typically comprises two nucleoside/nucleotide analog reverse transcriptase (RT) inhibitors (NRTIs) in combination with a non-nucleoside RT inhibitor or a protease inhibitor. Unfortunately, HIV is capable of escaping the therapy by mutating into drug-resistant variants. Computational models that correlate HIV drug susceptibilities to the virus genotype and to drug molecular properties might facilitate selection of improved combination treatment regimens. METHODOLOGY/PRINCIPAL FINDINGS: We applied our earlier developed proteochemometric modeling technology to analyze HIV mutant susceptibility to the eight clinically approved NRTIs. The data set used covered 728 virus variants genotyped for 240 sequence residues of the DNA polymerase domain of the RT; 165 of these residues contained mutations; totally the data-set covered susceptibility data for 4,495 inhibitor-RT combinations. Inhibitors and RT sequences were represented numerically by 3D-structural and physicochemical property descriptors, respectively. The two sets of descriptors and their derived cross-terms were correlated to the susceptibility data by partial least-squares projections to latent structures. The model identified more than ten frequently occurring mutations, each conferring more than two-fold loss of susceptibility for one or several NRTIs. The most deleterious mutations were K65R, Q151M, M184V/I, and T215Y/F, each of them decreasing susceptibility to most of the NRTIs. The predictive ability of the model was estimated by cross-validation and by external predictions for new HIV variants; both procedures showed very high correlation between the predicted and actual susceptibility values (Q2=0.89 and Q2ext=0.86). The model is available at www.hivdrc.org as a free web service for the prediction of the susceptibility to any of the clinically used NRTIs for any HIV-1 mutant variant. CONCLUSIONS/SIGNIFICANCE: Our results give directions how to develop approaches for selection of genome-based optimum combination therapy for patients harboring mutated HIV variants.

Toxicology has been an active research field for many decades, with academic, industrial and government involvement. Modern omics and computational approaches are changing the field, from merely disease-specific observational models into target-specific predictive models. Traditionally, toxicology has strong links with other fields such as biology, chemistry, pharmacology and medicine. With the rise of synthetic and new engineered materials, alongside ongoing prioritisation needs in chemical risk assessment for existing chemicals, early predictive evaluations are becoming of utmost importance to both scientific and regulatory purposes. ELIXIR is an intergovernmental organisation that brings together life science resources from across Europe. To coordinate the linkage of various life science efforts around modern predictive toxicology, the establishment of a new ELIXIR Community is seen as instrumental. In the past few years, joint efforts, building on incidental overlap, have been piloted in the context of ELIXIR. For example, the EU-ToxRisk, diXa, HeCaToS, transQST, and the nanotoxicology community have worked with the ELIXIR TeSS, Bioschemas, and Compute Platforms and activities. In 2018, a core group of interested parties wrote a proposal, outlining a sketch of what this new ELIXIR Toxicology Community would look like. A recent workshop (held September 30th to October 1st, 2020) extended this into an ELIXIR Toxicology roadmap and a shortlist of limited investment-high gain collaborations to give body to this new community. This Whitepaper outlines the results of these efforts and defines our vision of the ELIXIR Toxicology Community and how it complements other ELIXIR activities.