MOTIVATION: Untargeted metabolomics involves a large-scale comparison of the fragmentation pattern of a mass spectrum against a database containing known spectra. Given the number of comparisons involved, this step can be time-consuming. RESULTS: In this work, we present a GPU-accelerated cosine similarity implementation for Tandem Mass Spectrometry (MS), with an approximately 1000-fold speedup compared to the MatchMS reference implementation, without any loss of accuracy. This improvement enables repository-scale spectral library matching for compound identification without the need for large compute clusters. This impact extends to any spectral comparison-based methods such as molecular networking approaches and analogue search. AVAILABILITY AND IMPLEMENTATION: All code, results, and notebooks supporting are freely available under the MIT license at https://github.com/pangeAI/simms/.

Many therapies in clinical trials are based on single drug-single target relationships. To further extend this concept to multi-target approaches using multi-targeted drugs, we developed a machine learning pipeline to unravel the target landscape of kinase inhibitors. This pipeline, which we call 3D-KINEssence, uses a new type of protein fingerprints (3D FP) based on the structure of kinases generated through a 3D convolutional neural network (3D-CNN). These 3D-CNN kinase fingerprints were matched to molecular Morgan fingerprints to predict the targets of each respective kinase inhibitor based on available bioactivity data. The performance of the pipeline was evaluated on two test sets: a sparse drug-target set where each drug is matched in most cases to a single target and also on a densely-covered drug-target set where each drug is matched to most if not all targets. This latter set is more challenging to train, given its non-exclusive character. Our model's root-mean-square error (RMSE) based on the two datasets was 0.68 and 0.8, respectively. These results indicate that 3D FP can predict the target landscape of kinase inhibitors at around 0.8 log units of bioactivity. Our strategy can be utilized in proteochemometric or chemogenomic workflows by consolidating the target landscape of kinase inhibitors.

Drug discovery and development is a complex and costly process. Machine learning approaches are being investigated to help improve the effectiveness and speed of multiple stages of the drug discovery pipeline. Of these, those that use Knowledge Graphs (KG) have promise in many tasks, including drug repurposing, drug toxicity prediction and target gene-disease prioritization. In a drug discovery KG, crucial elements including genes, diseases and drugs are represented as entities, while relationships between them indicate an interaction. However, to construct high-quality KGs, suitable data are required. In this review, we detail publicly available sources suitable for use in constructing drug discovery focused KGs. We aim to help guide machine learning and KG practitioners who are interested in applying new techniques to the drug discovery field, but who may be unfamiliar with the relevant data sources. The datasets are selected via strict criteria, categorized according to the primary type of information contained within and are considered based upon what information could be extracted to build a KG. We then present a comparative analysis of existing public drug discovery KGs and an evaluation of selected motivating case studies from the literature. Additionally, we raise numerous and unique challenges and issues associated with the domain and its datasets, while also highlighting key future research directions. We hope this review will motivate KGs use in solving key and emerging questions in the drug discovery domain.

Analyzing the relationships among various drugs is an essential issue in the field of computational biology. Different kinds of informative knowledge, such as drug repurposing, can be extracted from drug-drug relationships. Scientific literature represents a rich source for the retrieval of knowledge about the relationships between biological concepts, mainly drug-drug, disease-disease, and drug-disease relationships. In this paper, we propose DDREL as a general-purpose method that applies deep learning on scientific literature to automatically extract the graph of syntactic and semantic relationships among drugs. DDREL remarkably outperforms the existing human drug network method and a random network respected to average similarities of drugs’ anatomical therapeutic chemical (ATC) codes. DDREL is able to shed light on the existing deficiency of the ATC codes in various drug groups. From the DDREL graph, the history of drug discovery became visible. In addition, drugs that had repurposing score 1 (diflunisal, pargyline, fenofibrate, guanfacine, chlorzoxazone, doxazosin, oxymetholone, azathioprine, drotaverine, demecarium, omifensine, yohimbine) were already used in additional indication. The proposed DDREL method justifies the predictive power of textual data in PubMed abstracts. DDREL shows that such data can be used to 1- Predict repurposing drugs with high accuracy, and 2- Reveal existing deficiencies of the ATC codes in various drug groups.

MOTIVATION: Combining disease relationships across multiple biological levels could aid our understanding of common processes taking place in disease, potentially indicating opportunities for drug sharing. Here, we propose a similarity fusion approach which accounts for differences in information content between different data types, allowing combination of each data type in a balanced manner. RESULTS: We apply this method to six different types of biological data (ontological, phenotypic, literature co-occurrence, genetic association, gene expression and drug indication data) for 84 diseases to create a 'disease map': a network of diseases connected at one or more biological levels. As well as reconstructing known disease relationships, 15% of links in the disease map are novel links spanning traditional ontological classes, such as between psoriasis and inflammatory bowel disease. 62% of links in the disease map represent drug-sharing relationships, illustrating the relevance of the similarity fusion approach to the identification of potential therapeutic relationships. AVAILABILITY AND IMPLEMENTATION: Freely available under the MIT license at https://github.com/e-oerton/disease-similarity-fusion. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

High-throughput screening (HTS) campaigns are routinely performed in pharmaceutical companies to explore activity profiles of chemical libraries for the identification of promising candidates for further investigation. With the aim of improving hit rates in these campaigns, data-driven approaches have been used to design relevant compound screening collections, enable effective hit triage and perform activity modeling for compound prioritization. Remarkable progress has been made in the activity modeling area since the recent introduction of large-scale bioactivity-based compound similarity metrics. This is evidenced by increased hit rates in iterative screening strategies and novel insights into compound mode of action obtained through activity modeling. Here, we provide an overview of the developments in data-driven approaches, elaborate on novel activity modeling techniques and screening paradigms explored and outline their significance in HTS.

Summary: In this work, we present eMolTox, a web server for the prediction of potential toxicity associated with a given molecule. A total of 174 toxicology-related in vitro/vivo experimental datasets were used for model construction and Mondrian conformal prediction was used to estimate the confidence of the resulting predictions. Toxic substructure analysis is also implemented in eMolTox. eMolTox predicts and displays a wealth of information of potential molecular toxicities for safety analysis in drug development. Availability and implementation: The eMolTox Server is freely available for use on the web at http://xundrug.cn/moltox. Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: In silico approaches often fail to utilize bioactivity data available for orthologous targets due to insufficient evidence highlighting the benefit for such an approach. Deeper investigation into orthologue chemical space and its influence toward expanding compound and target coverage is necessary to improve the confidence in this practice. Results: Here we present analysis of the orthologue chemical space in ChEMBL and PubChem and its impact on target prediction. We highlight the number of conflicting bioactivities between human and orthologues is low and annotations are overall compatible. Chemical space analysis shows orthologues are chemically dissimilar to human with high intra-group similarity, suggesting they could effectively extend the chemical space modelled. Based on these observations, we show the benefit of orthologue inclusion in terms of novel target coverage. We also benchmarked predictive models using a time-series split and also using bioactivities from Chemistry Connect and HTS data available at AstraZeneca, showing that orthologue bioactivity inclusion statistically improved performance. Availability and implementation: Orthologue-based bioactivity prediction and the compound training set are available at www.github.com/lhm30/PIDGINv2. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: While drug combination therapies are a well-established concept in cancer treatment, identifying novel synergistic combinations is challenging due to the size of combinatorial space. However, computational approaches have emerged as a time- and cost-efficient way to prioritize combinations to test, based on recently available large-scale combination screening data. Recently, Deep Learning has had an impact in many research areas by achieving new state-of-the-art model performance. However, Deep Learning has not yet been applied to drug synergy prediction, which is the approach we present here, termed DeepSynergy. DeepSynergy uses chemical and genomic information as input information, a normalization strategy to account for input data heterogeneity, and conical layers to model drug synergies. Results: DeepSynergy was compared to other machine learning methods such as Gradient Boosting Machines, Random Forests, Support Vector Machines and Elastic Nets on the largest publicly available synergy dataset with respect to mean squared error. DeepSynergy significantly outperformed the other methods with an improvement of 7.2% over the second best method at the prediction of novel drug combinations within the space of explored drugs and cell lines. At this task, the mean Pearson correlation coefficient between the measured and the predicted values of DeepSynergy was 0.73. Applying DeepSynergy for classification of these novel drug combinations resulted in a high predictive performance of an AUC of 0.90. Furthermore, we found that all compared methods exhibit low predictive performance when extrapolating to unexplored drugs or cell lines, which we suggest is due to limitations in the size and diversity of the dataset. We envision that DeepSynergy could be a valuable tool for selecting novel synergistic drug combinations. Availability and implementation: DeepSynergy is available via www.bioinf.jku.at/software/DeepSynergy. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

MOTIVATION: Recent large-scale omics initiatives have catalogued the somatic alterations of cancer cell line panels along with their pharmacological response to hundreds of compounds. In this study, we have explored these data to advance computational approaches that enable more effective and targeted use of current and future anticancer therapeutics. RESULTS: We modelled the 50% growth inhibition bioassay end-point (GI50) of 17,142 compounds screened against 59 cancer cell lines from the NCI60 panel (941,831 data-points, matrix 93.08% complete) by integrating the chemical and biological (cell line) information. We determine that the protein, gene transcript and miRNA abundance provide the highest predictive signal when modelling the GI50 endpoint, which significantly outperformed the DNA copy-number variation or exome sequencing data (Tukey's Honestly Significant Difference, P <0.05). We demonstrate that, within the limits of the data, our approach exhibits the ability to both interpolate and extrapolate compound bioactivities to new cell lines and tissues and, although to a lesser extent, to dissimilar compounds. Moreover, our approach outperforms previous models generated on the GDSC dataset. Finally, we determine that in the cases investigated in more detail, the predicted drug-pathway associations and growth inhibition patterns are mostly consistent with the experimental data, which also suggests the possibility of identifying genomic markers of drug sensitivity for novel compounds on novel cell lines. CONTACT: [email protected]; [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Analyzing different drugs for various purposes is an important issue in the area of computational biology. We categorize the previous computational studies into Individual and Network approaches. While the Individual approach focuses on one specific drug without considering its relationship with ot her drugs, the Network approach considers also the drugs relationships. In this paper, we apply a Network approach, previously proposed for discovering the relationships among diseases, to drug data. We construct a Human Drug Network (HDN) for 200 different drugs based on functional and structural information available in the PPI network. For evaluating our proposed HDN, first, we analyzed the literature to prove that the proposed HDN is biologically meaningful. Second, we used the HDN to augment the initial prior knowledge of different drugs. As an example of prior knowledge, we considered the initial seed proteins (a set of proteins which are previously known to be drug targets) of each drug. We clustered the HDN nodes using the Markov CLustering Algorithm (MCL) and then, we augmented the seed proteins of each drug based on the cluster it belongs to. In the end, we concluded that our proposed HDN enables us to generate novel hypotheses (in terms of potential drug target proteins) and produce complementary results comparing to existing methods.

Proteochemometrics (PCM) is a predictive bioactivity modelling method which simultaneously models the bioactivity of multiple ligands against multiple targets. PCM permits exploration of the selectivity and promiscuity of ligands on biomolecular systems of different complexity. This includes proteins and even cell-line models [ 1 , 2 ]. The suitability of PCM to predict compound polypharmacology has been validated both retrospectively and in prospective experimental validation [ 1 , 2 ]. In practice, each ligand-target interaction is encoded by the concatenation of ligand and target descriptor vectors used to train a single machine learning model. The inclusion of both chemical and target information enables the extra- and interpolation on the chemical and on the biological space. Therefore, PCM permits to predict compound bioactivities on targets not present in the training phase [ 3 ]. In this contribution, we show a methodological advancement in the field [ 4 ], namely how Bayesian inference (Gaussian Processes) can be successfully applied in the context of PCM for (i) the prediction of compound bioactivity along with the error estimation of the prediction; (ii) the determination of the applicability domain of a PCM model; and (iii) the inclusion of experimental uncertainty of bioactivity measurements. We illustrate how the application of PCM can be useful in medicinal chemistry to concomitantly optimize compounds selectivity and potency, in the context of two application scenarios: (a) modelling isoform-selective cyclooxygenase inhibition; and (b) large-scale cancer cell line drug sensitivity prediction, where we benchmark the predictive signal of basal gene expression, gene copy-number variation, exome sequencing, and protein abundance data. We present the R package C hemically A ware M odel B uilder ( camb ) [ 5 ], which is able to perform the above mentioned modelling tasks. camb is an open source platform for the generation of Structure-Activity and Structure-Property models. The functionalities of camb include: (i) standardisation of chemical structure representation, (ii) calculation of 905 one-dimensional descriptors and 14 fingerprints for small molecules, (iii) 8 types of amino acid descriptors, (iv) 13 whole protein sequence descriptors, and (iv) training, validation and visualization of predictive models. Overall, the application of PCM in these two case scenarios let us conclude that PCM is a suitable technique, on this data, to model the activity of ligands exhibiting diverse bioactivity profiles across a panel of targets, which can range from protein binding sites (a), to cancer cell-lines (b). The camb package constitutes a platform encompassing all steps for the generation of predictive models from chemical structures and their associated bioactivities/properties, which will provide reproducibility and simplify the generation of predictive bioactivity/property models.

The cellular metabolism of a living organism is among the most complex systems that man is currently trying to understand. Part of it is described by so-called protein-protein interaction (PPI) networks, and much effort is spent on analyzing these networks. Recently, there has been much interest in predicting involvement of network nodes (in this case, proteins) in different diseases. Many approaches to this problem exist. We categorize the previous studies into Individual and Network approaches. While the Individual approach focuses on one specific disease without considering its relationship with other diseases, the Network approach considers also these relationships. In this paper, we construct a Human Disease Network (HDN), using a novel approach for discovering relationships among different diseases. We built the HDN for 20 different diseases based on functional and structural information available in the PPI network. We showed that the proposed HDN is biologically meaningful and is capable of augmenting the initial prior knowledge of different diseases by sharing information across highly-related diseases. Furthermore, comparing to previous Individual and Network approaches, our proposed HDN increases the accuracy of predictive models and discovers more and still informative relationships among different diseases, respectively.

Infection with HIV cannot currently be cured; however it can be controlled by combination treatment with multiple anti-retroviral drugs. Given different viral genotypes for virtually each individual patient, the question now arises which drug combination to use to achieve effective treatment. With the availability of viral genotypic data and clinical phenotypic data, it has become possible to create computational models able to predict an optimal treatment regimen for an individual patient. Current models are based only on sequence data derived from viral genotyping; chemical similarity of drugs is not considered. To explore the added value of chemical similarity inclusion we applied proteochemometric models, combining chemical and protein target properties in a single bioactivity model. Our dataset was a large scale clinical database of genotypic and phenotypic information (in total ca. 300,000 drug-mutant bioactivity data points, 4 (NNRTI), 8 (NRTI) or 9 (PI) drugs, and 10,700 (NNRTI) 10,500 (NRTI) or 27,000 (PI) mutants). Our models achieved a prediction error below 0.5 Log Fold Change. Moreover, when directly compared with previously published sequence data, derived models PCM performed better in resistance classification and prediction of Log Fold Change (0.76 log units versus 0.91). Furthermore, we were able to successfully confirm both known and identify previously unpublished, resistance-conferring mutations of HIV Reverse Transcriptase (e.g. K102Y, T216M) and HIV Protease (e.g. Q18N, N88G) from our dataset. Finally, we applied our models prospectively to the public HIV resistance database from Stanford University obtaining a correct resistance prediction rate of 84% on the full set (compared to 80% in previous work on a high quality subset). We conclude that proteochemometric models are able to accurately predict the phenotypic resistance based on genotypic data even for novel mutants and mixtures. Furthermore, we add an applicability domain to the prediction, informing the user about the reliability of predictions.

BACKGROUND: G protein-coupled receptors (GPCRs) represent a family of well-characterized drug targets with significant therapeutic value. Phylogenetic classifications may help to understand the characteristics of individual GPCRs and their subtypes. Previous phylogenetic classifications were all based on the sequences of receptors, adding only minor information about the ligand binding properties of the receptors. In this work, we compare a sequence-based classification of receptors to a ligand-based classification of the same group of receptors, and evaluate the potential to use sequence relatedness as a predictor for ligand interactions thus aiding the quest for ligands of orphan receptors. RESULTS: We present a classification of GPCRs that is purely based on their ligands, complementing sequence-based phylogenetic classifications of these receptors. Targets were hierarchically classified into phylogenetic trees, for both sequence space and ligand (substructure) space. The overall organization of the sequence-based tree and substructure-based tree was similar; in particular, the adenosine receptors cluster together as well as most peptide receptor subtypes (e.g. opioid, somatostatin) and adrenoceptor subtypes. In ligand space, the prostanoid and cannabinoid receptors are more distant from the other targets, whereas the tachykinin receptors, the oxytocin receptor, and serotonin receptors are closer to the other targets, which is indicative for ligand promiscuity. In 93% of the receptors studied, de-orphanization of a simulated orphan receptor using the ligands of related receptors performed better than random (AUC > 0.5) and for 35% of receptors de-orphanization performance was good (AUC > 0.7). CONCLUSIONS: We constructed a phylogenetic classification of GPCRs that is solely based on the ligands of these receptors. The similarities and differences with traditional sequence-based classifications were investigated: our ligand-based classification uncovers relationships among GPCRs that are not apparent from the sequence-based classification. This will shed light on potential cross-reactivity of GPCR ligands and will aid the design of new ligands with the desired activity profiles. In addition, we linked the ligand-based classification with a ligand-focused sequence-based classification described in literature and proved the potential of this method for de-orphanization of GPCRs.

There exist several applications of multi-objective evolutionary algorithms for drug design, however, a common drawback in recent approaches is that the diversity of resulting molecule populations is relatively low. This paper seeks to overcome this problem by introducing niching as a technique to enhance search space diversity. A single population approach with dynamic niche identification is studied in the application domain. In order to apply niching in molecular spaces a metric for measuring the dissimilarity of molecules will be introduced. The approach will be validated in case studies and compared with results of an NSGA-II algorithm without niching in the search space.