Understanding Cytotoxicity and Cytostaticity in a High-Throughput Screening Collection

Proteomic Characterization of Undifferentiated Small Round Cell Sarcomas with EWSR1- and CIC::DUX4-Translocations Reveals Diverging Tumor Biology and Distinct Diagnostic Markers

Undifferentiated small round cell sarcomas of bone and soft tissue (USRS) are a group of tumors with heterogenic genomic alterations sharing similar morphology. In the present study, we performed a comparative large-scale proteomic analysis of USRS (n=42) with diverse genomic translocations including classic Ewing sarcomas with EWSR1::FLI1 fusions (n=24) or EWSR1::ERG - fusions (n=4), sarcomas with an EWSR1 - rearrangement (n=2), CIC::DUX4 fusion (n=8), as well as tumors classified as USRS with no genetic data available (n=4). Proteins extracted from formalin-fixed, paraffin-embedded (FFPE) pretherapeutic biopsies were analyzed qualitatively and quantitatively using shot gun mass spectrometry (MS). More than 8000 protein groups could be quantified using data-independent acquisition. Unsupervised hierarchical cluster analysis based on proteomic data allowed stratification of the 42 cases into distinct groups reflecting the different molecular genotypes. Protein signatures that significantly correlated with the respective genomic translocations were identified and used to generate a heatmap of all 42 sarcomas with assignment of cases with unknown molecular genetic data to either the EWSR1- or CIC-rearranged groups. MS-based prediction of sarcoma subtypes was molecularly confirmed in two cases where next-generation sequencing was technically feasible. MS also detected proteins routinely used in the immunohistochemical approach for the differential diagnosis of USRS. BCL11B highly expressed in Ewing sarcomas and Bach2 as well as ETS-1 highly expressed in CIC::DUX4-associated sarcomas, were among proteins identified by the present proteomic study and were chosen for immunohistochemical confirmation of MS data in our study cohort. Differential expression of these 3 markers in the two genetic groups were further validated in an independent cohort of n= 34 USRS. Finally, our proteomic results point towards diverging signaling pathways in the different USRS subgroups.

A guide to the use of bioassays in exploration of natural resources

Bioassays are the main tool to decipher bioactivities from natural resources thus their selection and quality are critical for optimal bioprospecting. They are used both in the early stages of compounds isolation/purification/identification, and in later stages to evaluate their safety and efficacy. In this review, we provide a comprehensive overview of the most common bioassays used in the discovery and development of new bioactive compounds with a focus on marine bioresources. We present a comprehensive list of practical considerations for selecting appropriate bioassays and discuss in detail the bioassays typically used to explore antimicrobial, antibiofilm, cytotoxic, antiviral, antioxidant, and anti-ageing potential. The concept of quality control and bioassay validation are introduced, followed by safety considerations, which are critical to advancing bioactive compounds to a higher stage of development. We conclude by providing an application-oriented view focused on the development of pharmaceuticals, food supplements, and cosmetics, the industrial pipelines where currently known marine natural products hold most potential. We highlight the importance of gaining reliable bioassay results, as these serve as a starting point for application-based development and further testing, as well as for consideration by regulatory authorities.

The cytotoxic and antiproliferative properties of ruthenium nitrosyl complexes and their modulation effect on cytochrome P450 in the HepG2 cell line

Ruthenium nitrosyl complexes are actively investigated as antitumor agents. Evaluation of potential interactions between cytochromes P450 (CYPs) with new compounds is carried out regularly during early drug development. In this study we have investigated the cytotoxic and antiproliferative activities of ruthenium nitrosyl complexes with methyl/ethyl esters of nicotinic and isonicotinic acids and γ-picoline against 2D and 3D cultures of human hepatocellular carcinoma HepG2 and non-cancer human lung fibroblasts MRC-5, assessed their photoinduced activity at λrad = 445 nm, and also evaluated their modulating effect on CYP3A4, CYP2C9, and CYP2C19. The study of cytotoxic and antiproliferative activities against 2D and 3D cell models was performed using phenotypic-based high content screening (HCS). The expression of CYP3A4, CYP2C9, and CYP2C19 mRNAs and CYP3A4 protein was examined using target-based HCS. The results of CYP3A4 mRNA expression were confirmed by real-time reverse transcription-polymerase chain reaction (RT-PCR). The ruthenium nitrosyl complexes exhibited a dose-dependent cytotoxic effect against HepG2 and MRC-5 cells. The cytotoxic activity of complexes with ethyl isonicotinate (1) and nicotinate (3, 4) was significantly lower for MRC-5 than for HepG2, for a complex with methyl isonicotinate (2) it was higher for MRC-5 than for HepG2, for a complex with γ-picoline (5) it was comparable for both lines. The antiproliferative effect of complexes 2 and 5 was one order of magnitude higher for MRC-5; for complexes 1, 3, and 4 it was comparable for both lines. The cytotoxic activity of all compounds for 3D HepG2 was lower than for 2D HepG2, with the exception of 4. Photoactivation affected the activity of complex 1 only. Its cytotoxic activity decreased, while the antiproliferative activity increased. The ruthenium nitrosyl complexes 1-4 acted as inducers of CYP3A4 and CYP2C19, while the complex with γ-picoline (5) induced of CYP3A4. Among the studied ruthenium nitrosyl complexes, the most promising potential antitumor compound is the ruthenium compound with methyl nicotinate (4).

Review of machine learning and deep learning models for toxicity prediction

The ever-increasing number of chemicals has raised public concerns due to their adverse effects on human health and the environment. To protect public health and the environment, it is critical to assess the toxicity of these chemicals. Traditional in vitro and in vivo toxicity assays are complicated, costly, and time-consuming and may face ethical issues. These constraints raise the need for alternative methods for assessing the toxicity of chemicals. Recently, due to the advancement of machine learning algorithms and the increase in computational power, many toxicity prediction models have been developed using various machine learning and deep learning algorithms such as support vector machine, random forest, k-nearest neighbors, ensemble learning, and deep neural network. This review summarizes the machine learning- and deep learning-based toxicity prediction models developed in recent years. Support vector machine and random forest are the most popular machine learning algorithms, and hepatotoxicity, cardiotoxicity, and carcinogenicity are the frequently modeled toxicity endpoints in predictive toxicology. It is known that datasets impact model performance. The quality of datasets used in the development of toxicity prediction models using machine learning and deep learning is vital to the performance of the developed models. The different toxicity assignments for the same chemicals among different datasets of the same type of toxicity have been observed, indicating benchmarking datasets is needed for developing reliable toxicity prediction models using machine learning and deep learning algorithms. This review provides insights into current machine learning models in predictive toxicology, which are expected to promote the development and application of toxicity prediction models in the future.

Potential antitumor activity of garlic against colorectal cancer: focus on the molecular mechanisms of action

PURPOSE

The aim of this review is to highlight the potential of garlic phytoconstituents as antitumor agents in colorectal cancer management based on their molecular mechanisms of action, while asking if their consumption, as part of the human diet, might contribute to the prevention of colorectal cancer.

METHODS

To gather information on appropriate in vitro, in vivo and human observational studies on this topic, the keywords "Allium sativum", "garlic", "colorectal cancer", "antitumor effect", "in vitro", "in vivo", "garlic consumption" and "colorectal cancer risk" were searched in different combinations in the international databases ScienceDirect, PubMed and Google Scholar. After duplicate and reviews removal, 61 research articles and meta-analyses published between 2000 and 2022 in peer-reviewed journals were found and included in this review.

RESULTS

Garlic (Allium sativum) proves to be a rich source of compounds with antitumor potential. Garlic-derived extracts and several of its individual constituents, especially organosulfur compounds such as allicin, diallyl sulfide, diallyl disulfide, diallyl trisulfide, diallyl tetrasulfide, allylmethylsulfide, S-allylmercaptocysteine, Z-ajoene, thiacremonone and Se-methyl-L-selenocysteine were found to possess cytotoxic, cytostatic, antiangiogenic and antimetastatic activities in different in vitro and in vivo models of colorectal cancer. The molecular mechanisms for their antitumor effects are associated with the modulation of several well-known signaling pathways involved in cell cycle progression, especially G1-S and G2-M transitions, as well as both the intrinsic and extrinsic apoptotic pathways. However, even though in various animal models some of these compounds have chemopreventive effects, based on different human observational studies, a diet rich in garlic is not consistently associated with a lower risk of developing colorectal cancer.

CONCLUSION

Independent of the impact of garlic consumption on colorectal cancer initiation and promotion in humans, its constituents might be good candidates for future conventional and/or complementary therapies, based on their diverse mechanisms of action.

CellDeathPred: a deep learning framework for ferroptosis and apoptosis prediction based on cell painting

Cell death, such as apoptosis and ferroptosis, play essential roles in the process of development, homeostasis, and pathogenesis of acute and chronic diseases. The increasing number of studies investigating cell death types in various diseases, particularly cancer and degenerative diseases, has raised hopes for their modulation in disease therapies. However, identifying the presence of a particular cell death type is not an obvious task, as it requires computationally intensive work and costly experimental assays. To address this challenge, we present CellDeathPred, a novel deep-learning framework that uses high-content imaging based on cell painting to distinguish cells undergoing ferroptosis or apoptosis from healthy cells. In particular, we incorporate a deep neural network that effectively embeds microscopic images into a representative and discriminative latent space, classifies the learned embedding into cell death modalities, and optimizes the whole learning using the supervised contrastive loss function. We assessed the efficacy of the proposed framework using cell painting microscopy data sets from human HT-1080 cells, where multiple inducers of ferroptosis and apoptosis were used to trigger cell death. Our model confidently separates ferroptotic and apoptotic cells from healthy controls, with an average accuracy of 95% on non-confocal data sets, supporting the capacity of the CellDeathPred framework for cell death discovery.

Evaluating point-prediction uncertainties in neural networks for protein-ligand binding prediction

Neural Network (NN) models provide potential to speed up the drug discovery process and reduce its failure rates. The success of NN models requires uncertainty quantification (UQ) as drug discovery explores chemical space beyond the training data distribution. Standard NN models do not provide uncertainty information. Some methods require changing the NN architecture or training procedure, limiting the selection of NN models. Moreover, predictive uncertainty can come from different sources. It is important to have the ability to separately model different types of predictive uncertainty, as the model can take assorted actions depending on the source of uncertainty. In this paper, we examine UQ methods that estimate different sources of predictive uncertainty for NN models aiming at protein-ligand binding prediction. We use our prior knowledge on chemical compounds to design the experiments. By utilizing a visualization method we create non-overlapping and chemically diverse partitions from a collection of chemical compounds. These partitions are used as training and test set splits to explore NN model uncertainty. We demonstrate how the uncertainties estimated by the selected methods describe different sources of uncertainty under different partitions and featurization schemes and the relationship to prediction error.

Interaction between Butyrate and Tumor Necrosis Factor α in Primary Rat Colonocytes

Butyrate, a short-chain fatty acid, is utilized by the gut epithelium as energy and it improves the gut epithelial barrier. More recently, it has been associated with beneficial effects on immune and cardiovascular homeostasis. Conversely, tumor necrosis factor alpha (TNFα) is a pro-inflammatory and pro-hypertensive cytokine. While butyrate and TNFα are both linked with hypertension, studies have not yet addressed their interaction in the colon. Here, we investigated the capacity of butyrate to modulate a host of effects of TNFα in primary rodent colonic cells in vitro. We measured ATP levels, cell viability, mitochondrial membrane potential (MMP), reactive oxygen species (ROS), mitochondrial oxidative phosphorylation, and glycolytic activity in colonocytes following exposure to either butyrate or TNFα, or both. To address the potential mechanisms, transcripts related to oxidative stress, cell fate, and cell metabolism (Pdk1, Pdk2, Pdk4, Spr, Slc16a1, Slc16a3, Ppargc1a, Cs, Lgr5, Casp3, Tnfr2, Bax, Bcl2, Sod1, Sod2, and Cat) were measured, and untargeted liquid chromatography-tandem mass spectrometry (LC-MS/MS) was employed to profile the metabolic responses of colonocytes following exposure to butyrate and TNFα. We found that both butyrate and TNFα lowered cellular ATP levels towards a quiescent cell energy phenotype, characterized by decreased oxygen consumption and extracellular acidification. Co-treatment with butyrate ameliorated TNFα-induced cytotoxicity and the reduction in cell viability. Butyrate also opposed the TNFα-mediated decrease in MMP and mitochondrial-to-intracellular calcium ratios, suggesting that butyrate may protect colonocytes against TNFα-induced cytotoxicity by decreasing mitochondrial calcium flux. The relative expression levels of pyruvate dehydrogenase kinase 4 (Pdk4) were increased via co-treatment of butyrate and TNFα, suggesting the synergistic inhibition of glycolysis. TNFα alone reduced the expression of monocarboxylate transporters slc16a1 and slc16a3, suggesting effects of TNFα on butyrate uptake into colonocytes. Of the 185 metabolites that were detected with LC-MS, the TNFα-induced increase in biopterin produced the only significant change, suggesting an alteration in mitochondrial biogenesis in colonocytes. Considering the reports of elevated colonic TNFα and reduced butyrate metabolism in many conditions, including in hypertension, the present work sheds light on cellular interactions between TNFα and butyrate in colonocytes that may be important in understanding conditions of the colon.

An Extremely Low-Frequency Vortex Magnetic Field Modifies Protein Expression, Rearranges the Cytoskeleton, and Induces Apoptosis of a Human Neuroblastoma Cell Line

Homogeneous extremely low-frequency electromagnetic fields (ELF-EMFs) alter biological phenomena, including the cell phenotype and proliferation rate. Heterogenous vortex magnetic fields (VMFs), a new approach of exposure to magnetic fields, induce systematic movements on charged biomolecules from target cells; however, the effect of VMFs on living systems remains uncertain. Here, we designed, constructed, and characterized an ELF-VMF-modified Rodin's coil to expose SH-SY5Y cells. Samples were analyzed by performing 2D-differential-gel electrophoresis, identified by MALDI-TOF/TOF, validated by western blotting, and characterized by confocal microscopy. A total of 106 protein spots were differentially expressed; 40 spots were downregulated and 66 were upregulated in the exposed cell proteome, compared to the control cell proteome. The identified spots are associated with cytoskeleton and cell viability proteins, and according to the protein-protein interaction network, a significant interaction among them was found. Our data revealed a decrease in cell survival associated with apoptotic cells without effects on the cell cycle, as well as evident changes in the cytoskeleton. We demonstrated that ELF-VMFs, at a specific frequency and exposure time, alter the cell proteome and structurally affect the target cells. This is the first report showing that VMF application might be a versatile system for testing different hypotheses in living systems, using appropriate exposure parameters.© 2022 Bioelectromagnetics Society.

Comparison of Cellular Morphological Descriptors and Molecular Fingerprints for the Prediction of Cytotoxicity- and Proliferation-Related Assays

Cell morphology features, such as those from the Cell Painting assay, can be generated at relatively low costs and represent versatile biological descriptors of a system and thereby compound response. In this study, we explored cell morphology descriptors and molecular fingerprints, separately and in combination, for the prediction of cytotoxicity- and proliferation-related in vitro assay endpoints. We selected 135 compounds from the MoleculeNet ToxCast benchmark data set which were annotated with Cell Painting readouts, where the relatively small size of the data set is due to the overlap of required annotations. We trained Random Forest classification models using nested cross-validation and Cell Painting descriptors, Morgan and ErG fingerprints, and their combinations. While using leave-one-cluster-out cross-validation (with clusters based on physicochemical descriptors), models using Cell Painting descriptors achieved higher average performance over all assays (Balanced Accuracy of 0.65, Matthews Correlation Coefficient of 0.28, and AUC-ROC of 0.71) compared to models using ErG fingerprints (BA 0.55, MCC 0.09, and AUC-ROC 0.60) and Morgan fingerprints alone (BA 0.54, MCC 0.06, and AUC-ROC 0.56). While using random shuffle splits, the combination of Cell Painting descriptors with ErG and Morgan fingerprints further improved balanced accuracy on average by 8.9% (in 9 out of 12 assays) and 23.4% (in 8 out of 12 assays) compared to using only ErG and Morgan fingerprints, respectively. Regarding feature importance, Cell Painting descriptors related to nuclei texture, granularity of cells, and cytoplasm as well as cell neighbors and radial distributions were identified to be most contributing, which is plausible given the endpoint considered. We conclude that cell morphological descriptors contain complementary information to molecular fingerprints which can be used to improve the performance of predictive cytotoxicity models, in particular in areas of novel structural space.

Uncertainty quantification in drug design

Machine learning and artificial intelligence are increasingly being applied to the drug-design process as a result of the development of novel algorithms, growing access, the falling cost of computation and the development of novel technologies for generating chemically and biologically relevant data. There has been recent progress in fields such as molecular de novo generation, synthetic route prediction and, to some extent, property predictions. Despite this, most research in these fields has focused on improving the accuracy of the technologies, rather than on quantifying the uncertainty in the predictions. Uncertainty quantification will become a key component in autonomous decision making and will be crucial for integrating machine learning and chemistry automation to create an autonomous design-make-test-analyse cycle. This review covers the empirical, frequentist and Bayesian approaches to uncertainty quantification, and outlines how they can be used for drug design. We also outline the impact of uncertainty quantification on decision making.

A Novel Edge Effect Detection Method for Real-Time Cellular Analyzer Using Functional Principal Component Analysis

Real-time cellular analyzer (RTCA) has been generally applied to test the cytotoxicity of chemicals. However, several factors impact the experimental quality. A non-negligible factor is the abnormal time-dependent cellular response curves (TCRCs) of the wells located at the edge of the E-plate which is defined as edge effect. In this paper, a novel statistical analysis is proposed to detect the edge effect. First, TCRCs are considered as observations of a random variable in a functional space. Then, functional principal component analysis (FPCA) is adopted to extract the principal component (PC) functions of the TCRCs, and the first and second PCs of these curves are selected to distinguish abnormal TCRCs. The average TCRC of the inner wells with the same culture environment is set as the standard. If the distance between the scoring point of the standard curve and one designated scoring point exceeds the defined threshold, the corresponding TCRC of the designated point should be removed automatically. The experimental results demonstrate the effectiveness of the proposed algorithm. This method can be used as a standard method to resolve general time-dependent series issues.

Revealing cytotoxic substructures in molecules using deep learning

In drug development, late stage toxicity issues of a compound are the main cause of failure in clinical trials. In silico methods are therefore of high importance to guide the early design process to reduce time, costs and animal testing. Technical advances and the ever growing amount of available toxicity data enabled machine learning, especially neural networks, to impact the field of predictive toxicology. In this study, cytotoxicity prediction, one of the earliest handles in drug discovery, is investigated using a deep learning approach trained on a highly consistent in-house data set of over 34,000 compounds with a share of less than 5% of cytotoxic molecules. The model reached a balanced accuracy of over 70%, similar to previously reported studies using Random Forest. Albeit yielding good results, neural networks are often described as a black box lacking deeper mechanistic understanding of the underlying model. To overcome this absence of interpretability, a Deep Taylor Decomposition method is investigated to identify substructures that may be responsible for the cytotoxic effects, the so-called toxicophores. Furthermore, this study introduces cytotoxicity maps which provide a visual structural interpretation of the relevance of these substructures. Using this approach could be helpful in drug development to predict the potential toxicity of a compound as well as to generate new insights into the toxic mechanism. Moreover, it could also help to de-risk and optimize compounds.

Contrasting the impact of cytotoxic and cytostatic drug therapies on tumour progression

A tumour grows when the total division (birth) rate of its cells exceeds their total mortality (death) rate. The capability for uncontrolled growth within the host tissue is acquired via the accumulation of driver mutations which enable the tumour to progress through various hallmarks of cancer. We present a mathematical model of the penultimate stage in such a progression. We assume the tumour has reached the limit of its present growth potential due to cell competition that either results in total birth rate reduction or death rate increase. The tumour can then progress to the final stage by either seeding a metastasis or acquiring a driver mutation. We influence the ensuing evolutionary dynamics by cytotoxic (increasing death rate) or cytostatic (decreasing birth rate) therapy while keeping the effect of the therapy on net growth reduction constant. Comparing the treatments head to head we derive conditions for choosing optimal therapy. We quantify how the choice and the related gain of optimal therapy depends on driver mutation, metastasis, intrinsic cell birth and death rates, and the details of cell competition. We show that detailed understanding of the cell population dynamics could be exploited in choosing the right mode of treatment with substantial therapy gains.

Cells and organisms evolve to better survive in their environments and to adapt to new challenges. Such dynamics manifest in a particularly problematic way with the evolution of drug resistance, which is increasingly recognized as a key challenge for global health. Thus, developing therapy paradigms that factor in evolutionary dynamics is an important goal. Using a minimal mathematical model of a cancer cell population we contrast cytotoxic (increasing death rate) and cytostatic (decreasing birth rate) treatments while keeping the effect of the therapy on the net growth reduction constant. We then quantify how the choice and the related gain of optimal therapy depends on driver mutation, metastasis, intrinsic cell birth and death rates and the details of cell competition. Most importantly, we identify specific cell population dynamics under which a certain treatment could be significantly better than the alternative.

Leveraging heterogeneous data from GHS toxicity annotations, molecular and protein target descriptors and Tox21 assay readouts to predict and rationalise acute toxicity

Despite the increasing knowledge in both the chemical and biological domains the assimilation and exploration of heterogeneous datasets, encoding information about the chemical, bioactivity and phenotypic properties of compounds, remains a challenge due to requirement for overlap between chemicals assayed across the spaces. Here, we have constructed a novel dataset, larger than we have used in prior work, comprising 579 acute oral toxic compounds and 1427 non-toxic compounds derived from regulatory GHS information, along with their corresponding molecular and protein target descriptors and qHTS in vitro assay readouts from the Tox21 project. We found no clear association between the results of a FAFDrugs4 toxicophore screen and the acute oral toxicity classifications for our compound set; and a screen using a subset of the ToxAlerts toxicophores was also of limited utility, with only slight enrichment toward the toxic set (odds ratio of 1.48). We then investigated to what degree toxic and non-toxic compounds could be separated in each of the spaces, to compare their potential contribution to further analyses. Using an LDA projection, we found the largest degree of separation using chemical descriptors (Cohen’s d of 1.95) and the lowest degree of separation between toxicity classes using qHTS descriptors (Cohen’s d of 0.67). To compare the predictivity of the feature spaces for the toxicity endpoint, we next trained Random Forest (RF) acute oral toxicity classifiers on either molecular, protein target and qHTS descriptors. RFs trained on molecular and protein target descriptors were most predictive, with ROC AUC values of 0.80–0.92 and 0.70–0.85, respectively, across three test sets. RFs trained on both chemical and protein target descriptors combined exhibited similar predictive performance to the single-domain models (ROC AUC of 0.80–0.91). Model interpretability was improved by the inclusion of protein target descriptors, which allow the identification of specific targets (e.g. Retinal dehydrogenase) with literature links to toxic modes of action (e.g. oxidative stress). The dataset compiled in this study has been made available for future application.

Electrochemical DNA Sensors with Layered Polyaniline-DNA Coating for Detection of Specific DNA Interactions

A DNA sensor has been proposed on the platform of glassy carbon electrode modified with native DNA implemented between two electropolymerized layers of polyaniline. The surface layer was assembled by consecutive stages of potentiodynamic electrolysis, DNA drop casting, and second electrolysis, which was required for capsulation of the DNA molecules and prevented their leaching into the solution. Surface layer assembling was controlled by cyclic voltammetry, electrochemical impedance spectroscopy, atomic force, and scanning electron microscopy. For doxorubicin measurement, the DNA sensor was first incubated in the Methylene blue solution that amplified signal due to DNA intercalation and competition with the doxorubicin molecules for the DNA binding sites. The charge transfer resistance of the inner layer interface decreased with the doxorubicin concentration in the range from 1.0 pM to 0.1 μM (LOD 0.6 pM). The DNA sensor was tested for the analysis of spiked artificial urine samples and showed satisfactory recovery in concentration range of 0.05⁻10 μM. The DNA sensor developed can find application in testing of antitumor drugs and some other DNA damaging factors.

Next generation diversity-oriented synthesis: a paradigm shift from chemical diversity to biological diversity

Diversity-oriented synthesis adds biological performance as a new diversity element.

Protein structure and computational drug discovery

The first protein structures revealed a complex web of weak interactions stabilising the three-dimensional shape of the molecule. Small molecule ligands were then found to exploit these same weak binding events to modulate protein function or act as substrates in enzymatic reactions. As the understanding of ligand–protein binding grew, it became possible to firstly predict how and where a particular small molecule might interact with a protein, and then to identify putative ligands for a specific protein site. Computer-aided drug discovery, based on the structure of target proteins, is now a well-established technique that has produced several marketed drugs. We present here an overview of the various methodologies being used for structure-based computer-aided drug discovery and comment on possible future developments in the field.

Orthologue chemical space and its influence on target prediction

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

ab454@cam.ac.uk.

Supplementary information

Supplementary data are available at Bioinformatics online.

Naïve Bayesian Models for Vero Cell Cytotoxicity

PURPOSE

To advance translational research of potential therapeutic small molecules against infectious microbes, the compounds must display a relative lack of mammalian cell cytotoxicity. Vero cell cytotoxicity (CC₅₀) is a common initial assay for this metric. We explored the development of naïve Bayesian models that can enhance the probability of identifying non-cytotoxic compounds.

METHODS

Vero cell cytotoxicity assays were identified in PubChem, reformatted, and curated to create a training set with 8741 unique small molecules. These data were used to develop Bayesian classifiers, which were assessed with internal cross-validation, external tests with a set of 193 compounds from our laboratory, and independent validation with an additional diverse set of 1609 unique compounds from PubChem.

RESULTS

Evaluation with independent, external test and validation sets indicated that cytotoxicity Bayesian models constructed with the ECFP_6 descriptor were more accurate than those that used FCFP_6 fingerprints. The best cytotoxicity Bayesian model displayed predictive power in external evaluations, according to conventional and chance-corrected statistics, as well as enrichment factors.

CONCLUSIONS

The results from external tests demonstrate that our novel cytotoxicity Bayesian model displays sufficient predictive power to help guide translational research. To assist the chemical tool and drug discovery communities, our curated training set is being distributed as part of the Supplementary Material. Graphical Abstract Naive Bayesian models have been trained with publically available data and offer a useful tool for chemical biology and drug discovery to select for small molecules with a high probability of exhibiting acceptably low Vero cell cytotoxicity.

Extending in Silico Protein Target Prediction Models to Include Functional Effects

In silico protein target deconvolution is frequently used for mechanism-of-action investigations; however existing protocols usually do not predict compound functional effects, such as activation or inhibition, upon binding to their protein counterparts. This study is hence concerned with including functional effects in target prediction. To this end, we assimilated a bioactivity training set for 332 targets, comprising 817,239 active data points with unknown functional effect (binding data) and 20,761,260 inactive compounds, along with 226,045 activating and 1,032,439 inhibiting data points from functional screens. Chemical space analysis of the data first showed some separation between compound sets (binding and inhibiting compounds were more similar to each other than both binding and activating or activating and inhibiting compounds), providing a rationale for implementing functional prediction models. We employed three different architectures to predict functional response, ranging from simplistic random forest models ('Arch1') to cascaded models which use separate binding and functional effect classification steps ('Arch2' and 'Arch3'), differing in the way training sets were generated. Fivefold stratified cross-validation outlined cascading predictions provides superior precision and recall based on an internal test set. We next prospectively validated the architectures using a temporal set of 153,467 of in-house data points (after a 4-month interim from initial data extraction). Results outlined Arch3 performed with the highest target class averaged precision and recall scores of 71% and 53%, which we attribute to the use of inactive background sets. Distance-based applicability domain (AD) analysis outlined that Arch3 provides superior extrapolation into novel areas of chemical space, and thus based on the results presented here, propose as the most suitable architecture for the functional effect prediction of small molecules. We finally conclude including functional effects could provide vital insight in future studies, to annotate cases of unanticipated functional changeover, as outlined by our CHRM1 case study.

Identification of New Activators of Mitochondrial Fusion Reveals a Link between Mitochondrial Morphology and Pyrimidine Metabolism

Mitochondria are dynamic organelles that produce most of the cellular ATP, and are involved in many other cellular functions such as Ca²⁺ signaling, differentiation, apoptosis, cell cycle, and cell growth. One key process of mitochondrial dynamics is mitochondrial fusion, which is catalyzed by mitofusins (MFN1 and MFN2) and OPA1. The outer mitochondrial membrane protein MFN2 plays a relevant role in the maintenance of mitochondrial metabolism, insulin signaling, and mutations that cause neurodegenerative disorders. Therefore, modulation of proteins involved in mitochondrial dynamics has emerged as a potential pharmacological strategy. Here, we report the identification of small molecules by high-throughput screen that promote mitochondrial elongation in an MFN1/MFN2-dependent manner. Detailed analysis of their mode of action reveals a previously unknown connection between pyrimidine metabolism and mitochondrial dynamics. Our data indicate a link between pyrimidine biosynthesis and mitochondrial dynamics, which maintains cell survival under stress conditions characterized by loss of pyrimidine synthesis.

Identification of cytotoxic, glutathione-reactive moieties inducing accumulation of reactive oxygen species via glutathione depletion

Reactive oxygen species (ROS) play an essential role in the survival and progression of cancer. Moderate oxidative stress drives proliferation, whereas high levels of ROS induce cytotoxicity. Compared to cancer cells, healthy cells often exhibit lower levels of oxidative stress. Elevation of cellular ROS levels by small molecules could therefore induce cancer-specific cytotoxicity. We have employed high-throughput phenotypic screening to identify inducers of ROS accumulation. We found 4,5-dihalo-2-methylpyridazin-3-one (DHMP) and 2,3,4,5(6)-tetrachloro-6(5)-methylpyridine (TCMP) moieties to strongly deplete GSH, to cause ROS accumulation and to induce cell death. Small molecules containing these fragments will most likely share the same properties and should therefore be carefully considered in the development of bioactive molecules.

CHEMGENIE: integration of chemogenomics data for applications in chemical biology

Increasing amounts of biological data are accumulating in the pharmaceutical industry and academic institutions. However, data does not equal actionable information, and guidelines for appropriate data capture, harmonization, integration, mining, and visualization need to be established to fully harness its potential. Here, we describe ongoing efforts at Merck & Co. to structure data in the area of chemogenomics. We are integrating complementary data from both internal and external data sources into one chemogenomics database (Chemical Genetic Interaction Enterprise; CHEMGENIE). Here, we demonstrate how this well-curated database facilitates compound set design, tool compound selection, target deconvolution in phenotypic screening, and predictive model building.

Modelling compound cytotoxicity using conformal prediction and PubChem HTS data

The assessment of compound cytotoxicity is an important part of the drug discovery process. Accurate predictions of cytotoxicity have the potential to expedite decision making and save considerable time and effort. In this work we apply class conditional conformal prediction to model the cytotoxicity of compounds based on 16 high throughput cytotoxicity assays from PubChem. The data span 16 cell lines and comprise more than 440 000 unique compounds. The data sets are heavily imbalanced with only 0.8% of the tested compounds being cytotoxic. We trained one classification model for each cell line and validated the performance with respect to validity and accuracy. The generated models deliver high quality predictions for both toxic and non-toxic compounds despite the imbalance between the two classes. On external data collected from the same assay provider as one of the investigated cell lines the model had a sensitivity of 74% and a specificity of 65% at the 80% confidence level among the compounds assigned to a single class. Compared to previous approaches for large scale cytotoxicity modelling, this represents a balanced performance in the prediction of the toxic and non-toxic classes. The conformal prediction framework also allows the modeller to control the error frequency of the predictions, allowing predictions of cytotoxicity outcomes with confidence.