Advances in Activity Cliff Research

Chemoinformatic regression methods and their applicability domain

The growing interest in chemoinformatic model uncertainty calls for a summary of the most widely used regression techniques and how to estimate their reliability. Regression models learn a mapping from the space of explanatory variables to the space of continuous output values. Among other limitations, the predictive performance of the model is restricted by the training data used for model fitting. Identification of unusual objects by outlier detection methods can improve model performance. Additionally, proper model evaluation necessitates defining the limitations of the model, often called the applicability domain. Comparable to certain classifiers, some regression techniques come with built-in methods or augmentations to quantify their (un)certainty, while others rely on generic procedures. The theoretical background of their working principles and how to deduce specific and general definitions for their domain of applicability shall be explained.

Design, Synthesis, Biological Evaluation, and Molecular Docking Studies of Pleuromutilin Derivatives Containing Thiazole

In this study, we designed and synthesized a series of pleuromutilin derivatives containing thiazole. The in vitro antimicrobial efficacy of these synthesized compounds was examined by using four strains. Compared with tiamulin (MIC = 0.25 μg/mL), compound 14 exhibited potency in inhibiting MRSA growth (MIC = 0.0625 μg/mL) in these derivatives. Meanwhile, the time-killing kinetics further demonstrated that compound 14 could efficiently inhibit the MRSA growth. After exposure at 4 × MIC, the postantibiotic effect (PAE) of compound 14 was 1.29 h. Additionally, in thigh-infected mice, compound 14 exhibited a more potent antibacterial efficacy (-1.78 ± 0.28 log10 CFU/g) in reducing MRSA load compared to tiamulin (-1.21 ± 0.23 log10 CFU/g). Moreover, the MTT assay on RAW 264.7 cells demonstrated that compound 14 (8 μg/mL) had no significant cytotoxicity. Docking studies indicated the strong affinity of compound 14 toward the 50S ribosomal subunit, with a binding free energy of -9.63 kcal/mol. Taken together, it could be deduced that compound 14 was a promising candidate for treating MRSA infections.

OLB-AC: toward optimizing ligand bioactivities through deep graph learning and activity cliffs

MOTIVATION

Deep graph learning (DGL) has been widely employed in the realm of ligand-based virtual screening. Within this field, a key hurdle is the existence of activity cliffs (ACs), where minor chemical alterations can lead to significant changes in bioactivity. In response, several DGL models have been developed to enhance ligand bioactivity prediction in the presence of ACs. Yet, there remains a largely unexplored opportunity within ACs for optimizing ligand bioactivity, making it an area ripe for further investigation.

RESULTS

We present a novel approach to simultaneously predict and optimize ligand bioactivities through DGL and ACs (OLB-AC). OLB-AC possesses the capability to optimize ligand molecules located near ACs, providing a direct reference for optimizing ligand bioactivities with the matching of original ligands. To accomplish this, a novel attentive graph reconstruction neural network and ligand optimization scheme are proposed. Attentive graph reconstruction neural network reconstructs original ligands and optimizes them through adversarial representations derived from their bioactivity prediction process. Experimental results on nine drug targets reveal that out of the 667 molecules generated through OLB-AC optimization on datasets comprising 974 low-activity, noninhibitor, or highly toxic ligands, 49 are recognized as known highly active, inhibitor, or nontoxic ligands beyond the datasets' scope. The 27 out of 49 matched molecular pairs generated by OLB-AC reveal novel transformations not present in their training sets. The adversarial representations employed for ligand optimization originate from the gradients of bioactivity predictions. Therefore, we also assess OLB-AC's prediction accuracy across 33 different bioactivity datasets. Results show that OLB-AC achieves the best Pearson correlation coefficient (r2) on 27/33 datasets, with an average improvement of 7.2%-22.9% against the state-of-the-art bioactivity prediction methods.

AVAILABILITY AND IMPLEMENTATION

The code and dataset developed in this work are available at github.com/Yueming-Yin/OLB-AC.

Linking chemicals, genes and morphological perturbations to diseases

The progress in image-based high-content screening technology has facilitated high-throughput phenotypic profiling notably the quantification of cell morphology perturbation by chemicals. However, understanding the mechanism of action of a chemical and linking it to cell morphology and phenotypes remains a challenge in drug discovery. In this study, we intended to integrate molecules that induced transcriptomic perturbations and cellular morphological changes into a biological network in order to assess chemical-phenotypic relationships in humans. Such a network was enriched with existing disease information to suggest molecular and cellular profiles leading to phenotypes. Two datasets were used for this study. Firstly, we used the "Cell Painting morphological profiling assay" dataset, composed of 30,000 compounds tested on human osteosarcoma cells (named U2OS). Secondly, we used the "L1000 mRNA profiling assay" dataset, a collection of transcriptional expression data from cultured human cells treated with approximately 20,000 bioactive small molecules from the Library of Integrated Network-based Cellular Signatures (LINCS). Furthermore, pathways, gene ontology terms and disease enrichments were performed on the transcriptomics data. Overall, our study makes it possible to develop a biological network combining chemical-gene-pathway-morphological perturbation and disease relationships. It contains an ensemble of 9989 chemicals, 732 significant morphological features and 12,328 genes. Through diverse examples, we demonstrated that some drugs shared similar genes, pathways and morphological profiles that, taken together, could help in deciphering chemical-phenotype observations.

Computational Prediction of Drug Phenotypic Effects Based on Substructure-Phenotype Associations

Identifying drug phenotypic effects, including therapeutic effects and adverse drug reactions (ADRs), is an inseparable part for evaluating the potentiality of new drug candidates (NDCs). However, current computational methods for predicting phenotypic effects of NDCs are mainly based on the overall structure of an NDC or a related target. These approaches often lead to inconsistencies between the structures and functions and limit the prediction space of NDCs. In this study, first, we constructed quantitative associations of substructure-domain, domain-ADR, and domain-ATC (Anatomical Therapeutic Chemical Classification System code) through L1LOG and L1SVM machine learning models. These associations represent relationships between phenotypes (ADRs and ATCs) and local structures of drugs and proteins. Then, based on these established associations, substructure-phenotype relationships were constructed which were utilized to quantify drug-phenotype relationships. Thus, this approach could achieve high-throughput and effective evaluations of the druggability of NDCs by referring to the established substructure-phenotype relationships and structural information of NDCs without additional prior knowledge. Using this computational pipeline, 83,205 drug-ATC relationships (including 1,479 drugs and 178 ATCs) and 306,421 drug-ADR relationships (including 1,752 drugs and 454 ADRs) were predicted in total. The prediction results were validated at four levels: five-fold cross validation, public databases, literature, and molecular docking. Furthermore, three case studies demonstrated the feasibility of our method. 79 ATCs and 269 ADRs were predicted to be related to Maraviroc, an approved drug, including the existing antiviral effect in clinical use. Additionally, we also found risk substructures of severe ADRs, for example, SUB215 (>= 1, saturated or only aromatic carbon ring size 7) can result in shock. And we analyzed the mechanism of action (MOA) of interested drugs based on the established drug-substructure-domain-protein associations. In a word, this approach through establishing drug-substructure-phenotype relationships can achieve quantitative prediction of phenotypes for a given NDC or drug without any prior knowledge except its structure information. Using that way, we can directly obtain the relationships between substructure and phenotype of a compound, which is more convenient to analyze the phenotypic mechanism of drugs and accelerate the process of rational drug design.

Implications of Additivity and Nonadditivity for Machine Learning and Deep Learning Models in Drug Design

Matched molecular pairs (MMPs) are nowadays a commonly applied concept in drug design. They are used in many computational tools for structure-activity relationship analysis, biological activity prediction, or optimization of physicochemical properties. However, until now it has not been shown in a rigorous way that MMPs, that is, changing only one substituent between two molecules, can be predicted with higher accuracy and precision in contrast to any other chemical compound pair. It is expected that any model should be able to predict such a defined change with high accuracy and reasonable precision. In this study, we examine the predictability of four classical properties relevant for drug design ranging from simple physicochemical parameters (log D and solubility) to more complex cell-based ones (permeability and clearance), using different data sets and machine learning algorithms. Our study confirms that additive data are the easiest to predict, which highlights the importance of recognition of nonadditivity events and the challenging complexity of predicting properties in case of scaffold hopping. Despite deep learning being well suited to model nonlinear events, these methods do not seem to be an exception of this observation. Though they are in general performing better than classical machine learning methods, this leaves the field with a still standing challenge.

Development of binary classification models for assessment of drug-induced liver injury in humans using a large set of FDA-approved drugs

Drug-induced liver injury (DILI) has been identified as one of the major causes for drugs withdrawn from the market, and even termination during the late stages of development. Therefore, it is imperative to evaluate the DILI potential of lead compounds during the research and development process. Although various computational models have been developed to predict DILI, most of which applied the DILI data were extracted from preclinical sources. In this investigation, the in silico prediction models for DILI were constructed based on 1140 FDA-approved drugs by using naïve Bayes classifier approach. The genetic algorithm method was applied for the molecular descriptors selection. Among these established prediction models, the NB-11 model based on eight molecular descriptors combined with ECFP_18 showed the best prediction performance for DILI, which gave 91.7% overall prediction accuracy for the training set, and 68.9% concordance for the external test set. Therefore, the established NB-11 prediction model can be used as a reliable virtual screening tool to predict DILI adverse effect in the early stages of drug design. In addition, some new structural alters for DILI were identified, which could be used for structural optimization in the future drug design by medicinal chemists.

Deciphering a Pharmacophore Network: A Case Study Using BCR-ABL Data

This paper introduces a general method that can be used to create groups of pharmacophores to support their further in-depth analysis. A BCR-ABL molecular dataset was used to calculate graph edit distances between pharmacophores and led to their organization into a novel pharmacophore network. The application of a graph layout algorithm allowed us to discriminate between the pharmacophores associated with active compounds and those associated with inactive compounds. A clustering approach was used to refine the partitioning by grouping the pharmacophores based on their structures, activities, and binding modes. Analysis of a newly spatialized pharmacophore network provided us with critical insight into structure-activity relationships, most notably those that revealed distinctions between activity classes and chemical families. As shown, this method permits us to identify families of structurally homogeneous pharmacophores.

Nonadditivity in public and inhouse data: implications for drug design

Numerous ligand-based drug discovery projects are based on structure-activity relationship (SAR) analysis, such as Free-Wilson (FW) or matched molecular pair (MMP) analysis. Intrinsically they assume linearity and additivity of substituent contributions. These techniques are challenged by nonadditivity (NA) in protein-ligand binding where the change of two functional groups in one molecule results in much higher or lower activity than expected from the respective single changes. Identifying nonlinear cases and possible underlying explanations is crucial for a drug design project since it might influence which lead to follow. By systematically analyzing all AstraZeneca (AZ) inhouse compound data and publicly available ChEMBL25 bioactivity data, we show significant NA events in almost every second assay among the inhouse and once in every third assay in public data sets. Furthermore, 9.4% of all compounds of the AZ database and 5.1% from public sources display significant additivity shifts indicating important SAR features or fundamental measurement errors. Using NA data in combination with machine learning showed that nonadditive data is challenging to predict and even the addition of nonadditive data into training did not result in an increase in predictivity. Overall, NA analysis should be applied on a regular basis in many areas of computational chemistry and can further improve rational drug design.

FastTargetPred: a program enabling the fast prediction of putative protein targets for input chemical databases

SUMMARY

Several web-based tools predict the putative targets of a small molecule query compound by similarity to molecules with known bioactivity data using molecular fingerprints. In numerous situations, it would however be valuable to be able to run such computations on a local computer. We present FastTargetPred, a new program for the prediction of protein targets for small molecule queries. Structural similarity computations rely on a large collection of confirmed protein-ligand activities extracted from the curated ChEMBL 25 database. The program allows to annotate an input chemical library of ∼100k compounds within a few hours on a simple personal computer.

AVAILABILITY AND IMPLEMENTATION

FastTargetPred is written in Python 3 (≥3.7) and C languages. Python code depends only on the Python Standard Library. The program can be run on Linux, MacOS and Windows operating systems. Pre-compiled versions are available at https://github.com/ludovicchaput/FastTargetPred. FastTargetPred is licensed under the GNU GPLv3. The program calls some scripts from the free chemistry toolkit MayaChemTools.

CONTACT

bruno.villoutreix@inserm.fr.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Discovery of class I histone deacetylase inhibitors based on romidpesin with promising selectivity for cancer cells

Aim: Class I histone deacetylases (HDACs) are considered to be promising anticancer targets, but selective inhibition of class I HDAC isoforms remains a challenge. Methods & results: Previously, we obtained a selective class I HDAC inhibitor 9 based on a macrocyclic HDAC inhibitor Romidpesin. As our continuous efforts, a library of novel cyclicdepsipeptides based on 9 was established using a convergent synthesis strategy. The most active compounds 10, 16 and 19 selectively inhibit class I HDACs and exhibit promising nanomolar antiproliferative activities against several cancer cell lines with excellent selectivity toward cancer cells over normal cells. Besides, compound 10 demonstrates excellent antitumor effects in human prostate carcinoma PC3 xenograft models with no observed toxicity. Conclusion: These cyclicdepsipeptides show great therapeutic potential as novel anticancer agents for clinical translation.

Prediction of toxicity of secondary metabolites

The prediction of toxicological endpoints has gained broad acceptance; it is widely applied in early stages of drug discovery as well as for impurities obtained in the production of generic or equivalent products. In this work, we describe methodologies for the prediction of toxicological endpoints compounds, with a particular focus on secondary metabolites. Case studies include toxicity prediction of natural compound databases with anti-diabetic, anti-malaria and anti-HIV properties.

D3R Grand Challenge 3: blind prediction of protein-ligand poses and affinity rankings

The Drug Design Data Resource aims to test and advance the state of the art in protein-ligand modeling by holding community-wide blinded, prediction challenges. Here, we report on our third major round, Grand Challenge 3 (GC3). Held 2017-2018, GC3 centered on the protein Cathepsin S and the kinases VEGFR2, JAK2, p38-α, TIE2, and ABL1, and included both pose-prediction and affinity-ranking components. GC3 was structured much like the prior challenges GC2015 and GC2. First, Stage 1 tested pose prediction and affinity ranking methods; then all available crystal structures were released, and Stage 2 tested only affinity rankings, now in the context of the available structures. Unique to GC3 was the addition of a Stage 1b self-docking subchallenge, in which the protein coordinates from all of the cocrystal structures used in the cross-docking challenge were released, and participants were asked to predict the pose of CatS ligands using these newly released structures. We provide an overview of the outcomes and discuss insights into trends and best-practices.

A new chemoinformatics approach with improved strategies for effective predictions of potential drugs

Background

Fast and accurate identification of potential drug candidates against therapeutic targets (i.e., drug–target interactions, DTIs) is a fundamental step in the early drug discovery process. However, experimental determination of DTIs is time-consuming and costly, especially for testing the associations between the entire chemical and genomic spaces. Therefore, computationally efficient algorithms with accurate predictions are required to achieve such a challenging task. In this work, we design a new chemoinformatics approach derived from neighbor-based collaborative filtering (NBCF) to infer potential drug candidates for targets of interest. One of the fundamental steps of NBCF in the application of DTI predictions is to accurately measure the similarity between drugs solely based on the DTI profiles of known knowledge. However, commonly used similarity calculation methods such as COSINE may be noise-prone due to the extremely sparse property of the DTI bipartite network, which decreases the model performance of NBCF. We herein propose three strategies to remedy such a dilemma, which include: (1) adopting a positive pointwise mutual information (PPMI)-based similarity metric, which is noise-immune to some extent; (2) performing low-rank approximation of the original prediction scores; (3) incorporating auxiliary (complementary) information to produce the final predictions.

Results

We test the proposed methods in three benchmark datasets and the results indicate that our strategies are helpful to improve the NBCF performance for DTI predictions. Comparing to the prior algorithm, our methods exhibit better results assessed by a recall-based evaluation metric.

Conclusions

A new chemoinformatics approach with improved strategies was successfully developed to predict potential DTIs. Among them, the model based on the sparsity resistant PPMI similarity metric exhibits the best performance, which may be helpful to researchers for identifying potential drugs against therapeutic targets of interest, and can also be applied to related research such as identifying candidate disease genes.

Mathematical and Structural Characterization of Strong Nonadditive Structure-Activity Relationship Caused by Protein Conformational Changes

In medicinal chemistry, accurate prediction of additivity-based structure-activity relationship (SAR) analysis rests on three assumptions: (1) a consistent binding pose of the central scaffold, (2) no interaction between the R group substituents, and (3) a relatively rigid binding pocket in which the R group substituents act independently. Previously, examples of nonadditive SAR have been documented in systems that deviate from the first two assumptions. Local protein structural change upon ligand binding, through induced fit or conformational selection, although a well-known phenomenon that invalidates the third assumption, has not been linked to nonadditive SAR conclusively. Here, for the first time, we present clear structural evidence that the formation of a hydrophobic pocket upon ligand binding in PDE2 catalytic site reduces the size of another distinct subpocket and contributes to strong nonadditive SAR between two otherwise distant R groups.

Progress with modeling activity landscapes in drug discovery

INTRODUCTION

Activity landscapes (ALs) are representations and models of compound data sets annotated with a target-specific activity. In contrast to quantitative structure-activity relationship (QSAR) models, ALs aim at characterizing structure-activity relationships (SARs) on a large-scale level encompassing all active compounds for specific targets. The popularity of AL modeling has grown substantially with the public availability of large activity-annotated compound data sets. AL modeling crucially depends on molecular representations and similarity metrics used to assess structural similarity. Areas covered: The concepts of AL modeling are introduced and its basis in quantitatively assessing molecular similarity is discussed. The different types of AL modeling approaches are introduced. AL designs can broadly be divided into three categories: compound-pair based, dimensionality reduction, and network approaches. Recent developments for each of these categories are discussed focusing on the application of mathematical, statistical, and machine learning tools for AL modeling. AL modeling using chemical space networks is covered in more detail. Expert opinion: AL modeling has remained a largely descriptive approach for the analysis of SARs. Beyond mere visualization, the application of analytical tools from statistics, machine learning and network theory has aided in the sophistication of AL designs and provides a step forward in transforming ALs from descriptive to predictive tools. To this end, optimizing representations that encode activity relevant features of molecules might prove to be a crucial step.

ADMET Evaluation in Drug Discovery. Part 17: Development of Quantitative and Qualitative Prediction Models for Chemical-Induced Respiratory Toxicity

As a dangerous end point, respiratory toxicity can cause serious adverse health effects and even death. Meanwhile, it is a common and traditional issue in occupational and environmental protection. Pharmaceutical and chemical industries have a strong urge to develop precise and convenient computational tools to evaluate the respiratory toxicity of compounds as early as possible. Most of the reported theoretical models were developed based on the respiratory toxicity data sets with one single symptom, such as respiratory sensitization, and therefore these models may not afford reliable predictions for toxic compounds with other respiratory symptoms, such as pneumonia or rhinitis. Here, based on a diverse data set of mouse intraperitoneal respiratory toxicity characterized by multiple symptoms, a number of quantitative and qualitative predictions models with high reliability were developed by machine learning approaches. First, a four-tier dimension reduction strategy was employed to find an optimal set of 20 molecular descriptors for model building. Then, six machine learning approaches were used to develop the prediction models, including relevance vector machine (RVM), support vector machine (SVM), regularized random forest (RRF), extreme gradient boosting (XGBoost), naïve Bayes (NB), and linear discriminant analysis (LDA). Among all of the models, the SVM regression model shows the most accurate quantitative predictions for the test set (q²_ext = 0.707), and the XGBoost classification model achieves the most accurate qualitative predictions for the test set (MCC of 0.644, AUC of 0.893, and global accuracy of 82.62%). The application domains were analyzed, and all of the tested compounds fall within the application domain coverage. We also examined the structural features of the compounds and important fragments with large prediction errors. In conclusion, the SVM regression model and the XGBoost classification model can be employed as accurate prediction tools for respiratory toxicity.

Getting SMARt in drug discovery: chemoinformatics approaches for mining structure–multiple activity relationships

In light of the high relevance of polypharmacology, multi-target screening is a major trend in drug discovery.

Internet Databases of the Properties, Enzymatic Reactions, and Metabolism of Small Molecules-Search Options and Applications in Food Science

Internet databases of small molecules, their enzymatic reactions, and metabolism have emerged as useful tools in food science. Database searching is also introduced as part of chemistry or enzymology courses for food technology students. Such resources support the search for information about single compounds and facilitate the introduction of secondary analyses of large datasets. Information can be retrieved from databases by searching for the compound name or structure, annotating with the help of chemical codes or drawn using molecule editing software. Data mining options may be enhanced by navigating through a network of links and cross-links between databases. Exemplary databases reviewed in this article belong to two classes: tools concerning small molecules (including general and specialized databases annotating food components) and tools annotating enzymes and metabolism. Some problems associated with database application are also discussed. Data summarized in computer databases may be used for calculation of daily intake of bioactive compounds, prediction of metabolism of food components, and their biological activity as well as for prediction of interactions between food component and drugs.