PepExplainer: An explainable deep learning model for selection-based macrocyclic peptide bioactivity prediction and optimization.

Macrocyclic peptides possess unique features, making them highly promising as a drug modality. However, evaluating their bioactivity through wet lab experiments is generally resource-intensive and time-consuming. Despite advancements in artificial intelligence (AI) for bioactivity prediction, challenges remain due to limited data availability and the interpretability issues in deep learning models, often leading to less-than-ideal predictions. To address these challenges, we developed PepExplainer, an explainable graph neural network based on substructure mask explanation (SME). This model excels at deciphering amino acid substructures, translating macrocyclic peptides into detailed molecular graphs at the atomic level, and efficiently handling non-canonical amino acids and complex macrocyclic peptide structures. PepExplainer's effectiveness is enhanced by utilizing the correlation between peptide enrichment data from selection-based focused library and bioactivity data, and employing transfer learning to improve bioactivity predictions of macrocyclic peptides against IL-17C/IL-17 RE interaction. Additionally, PepExplainer underwent further validation for bioactivity prediction using an additional set of thirteen newly synthesized macrocyclic peptides. Moreover, it enabled the optimization of the IC50 of a macrocyclic peptide, reducing it from 15 nM to 5.6 nM based on the contribution score provided by PepExplainer. This achievement underscores PepExplainer's skill in deciphering complex molecular patterns, highlighting its potential to accelerate the discovery and optimization of macrocyclic peptides.

Machine Learning-Driven Classification of Urease Inhibitors Leveraging Physicochemical Properties as Effective Filter Criteria.

Urease, a pivotal enzyme in nitrogen metabolism, plays a crucial role in various microorganisms, including the pathogenic Helicobacter pylori. Inhibiting urease activity offers a promising approach to combating infections and associated ailments, such as chronic kidney diseases and gastric cancer. However, identifying potent urease inhibitors remains challenging due to resistance issues that hinder traditional approaches. Recently, machine learning (ML)-based models have demonstrated the ability to predict the bioactivity of molecules rapidly and effectively. In this study, we present ML models designed to predict urease inhibitors by leveraging essential physicochemical properties. The methodological approach involved constructing a dataset of urease inhibitors through an extensive literature search. Subsequently, these inhibitors were characterized based on physicochemical properties calculations. An exploratory data analysis was then conducted to identify and analyze critical features. Ultimately, 252 classification models were trained, utilizing a combination of seven ML algorithms, three attribute selection methods, and six different strategies for categorizing inhibitory activity. The investigation unveiled discernible trends distinguishing urease inhibitors from non-inhibitors. This differentiation enabled the identification of essential features that are crucial for precise classification. Through a comprehensive comparison of ML algorithms, tree-based methods like random forest, decision tree, and XGBoost exhibited superior performance. Additionally, incorporating the "chemical family type" attribute significantly enhanced model accuracy. Strategies involving a gray-zone categorization demonstrated marked improvements in predictive precision. This research underscores the transformative potential of ML in predicting urease inhibitors. The meticulous methodology outlined herein offers actionable insights for developing robust predictive models within biochemical systems.

Characterization and Classification In Silico of Peptides with Dual Activity (Antimicrobial and Wound Healing).

The growing challenge of chronic wounds and antibiotic resistance has spotlighted the potential of dual-function peptides (antimicrobial and wound healing) as novel therapeutic strategies. The investigation aimed to characterize and correlate in silico the physicochemical attributes of these peptides with their biological activity. We sourced a dataset of 207 such peptides from various peptide databases, followed by a detailed analysis of their physicochemical properties using bioinformatic tools. Utilizing statistical tools like clustering, correlation, and principal component analysis (PCA), patterns and relationships were discerned among these properties. Furthermore, we analyzed the peptides' functional domains for insights into their potential mechanisms of action. Our findings spotlight peptides in Cluster 2 as efficacious in wound healing, whereas Cluster 1 peptides exhibited pronounced antimicrobial potential. In our study, we identified specific amino acid patterns and peptide families associated with their biological activities, such as the cecropin antimicrobial domain. Additionally, we found the presence of polar amino acids like arginine, cysteine, and lysine, as well as apolar amino acids like glycine, isoleucine, and leucine. These characteristics are crucial for interactions with bacterial membranes and receptors involved in migration, proliferation, angiogenesis, and immunomodulation. While this study provides a groundwork for therapeutic development, translating these findings into practical applications necessitates additional experimental and clinical research.

Employing Molecular Conformations for Ligand-Based Virtual Screening with Equivariant Graph Neural Network and Deep Multiple Instance Learning.

Ligand-based virtual screening (LBVS) is a promising approach for rapid and low-cost screening of potentially bioactive molecules in the early stage of drug discovery. Compared with traditional similarity-based machine learning methods, deep learning frameworks for LBVS can more effectively extract high-order molecule structure representations from molecular fingerprints or structures. However, the 3D conformation of a molecule largely influences its bioactivity and physical properties, and has rarely been considered in previous deep learning-based LBVS methods. Moreover, the relative bioactivity benchmark dataset is still lacking. To address these issues, we introduce a novel end-to-end deep learning architecture trained from molecular conformers for LBVS. We first extracted molecule conformers from multiple public molecular bioactivity data and consolidated them into a large-scale bioactivity benchmark dataset, which totally includes millions of endpoints and molecules corresponding to 954 targets. Then, we devised a deep learning-based LBVS called EquiVS to learn molecule representations from conformers for bioactivity prediction. Specifically, graph convolutional network (GCN) and equivariant graph neural network (EGNN) are sequentially stacked to learn high-order molecule-level and conformer-level representations, followed with attention-based deep multiple-instance learning (MIL) to aggregate these representations and then predict the potential bioactivity for the query molecule on a given target. We conducted various experiments to validate the data quality of our benchmark dataset, and confirmed EquiVS achieved better performance compared with 10 traditional machine learning or deep learning-based LBVS methods. Further ablation studies demonstrate the significant contribution of molecular conformation for bioactivity prediction, as well as the reasonability and non-redundancy of deep learning architecture in EquiVS. Finally, a model interpretation case study on CDK2 shows the potential of EquiVS in optimal conformer discovery. The overall study shows that our proposed benchmark dataset and EquiVS method have promising prospects in virtual screening applications.

Machine Learning-Enabled Genome Mining and Bioactivity Prediction of Natural Products.

Natural products (NPs) produced by microorganisms and plants are a major source of drugs, herbicides, and fungicides. Thanks to recent advances in DNA sequencing, bioinformatics, and genome mining tools, a vast amount of data on NP biosynthesis has been generated over the years, which has been increasingly exploited to develop machine learning (ML) tools for NP discovery. In this review, we discuss the latest advances in developing and applying ML tools for exploring the potential NPs that can be encoded by genomic language and predicting the types of bioactivities of NPs. We also examine the technical challenges associated with the development and application of ML tools for NP research.

A consensual machine-learning-assisted QSAR model for effective bioactivity prediction of xanthine oxidase inhibitors using molecular fingerprints.

Xanthine oxidase inhibitors (XOIs) have been widely studied due to the promising potential as safe and effective therapeutics in hyperuricemia and gout. Currently, available XOI molecules have been developed from different experiments but they are with the wide structure diversity and significant varying bioactivities. So it is of great practical significance to present a consensual QSAR model for effective bioactivity prediction of XOIs based on a systematic compiling of these XOIs across different experiments. In this work, 249 XOIs belonging to 16 scaffolds were collected and were integrated into a consensual dataset by introducing the concept of IC50 values relative to allopurinol (RIC50). Here, extended connectivity fingerprints (ECFPs) were employed to represent XOI molecules. By performing effective feature selection by machine-learning method, 54 crucial fingerprints were indicated to be valuable for predicting the inhibitory potency (IP) of XOIs. The optimal predictor yields the promising performance by different cross-validation tests. Besides, an external validation of 43 XOIs and a case study on febuxostat also provide satisfactory results, indicating the powerful generalization of our predictor. Here, the predictor was interpreted by shapely additive explanation (SHAP) method which revealed several important substructures by mapping the featured fingerprints to molecular structures. Then, 15 new molecules were designed and predicted by our predictor to show superior IP than febuxostat. Finally, molecular docking simulation was performed to gain a deep insight into molecular binding mode with xanthine oxidase (XO) enzyme, showing that molecules with selenazole moiety, cyano group and isopropyl group tended to yield higher IP. The absorption, distribution, metabolism, excretion and toxicity (ADMET) prediction results further enhanced the potential of these novel XOIs as drug candidates. Overall, this work presents a QSAR model for accurate prediction of IP of XOIs, and is expected to provide new insights for further structure-guided design of novel XOIs.

Machine Learning Assisted Discovery of Novel p38α Inhibitors from Natural Products.

BACKGROUND: P38α, emerging as a hot spot for drug discovery, is a member of the mitogen- activated protein kinase (MAPK) family and plays a crucial role in regulating the production of inflammatory mediators. However, despite a massive number of highly potent molecules being reported and several under clinical trials, no p38α inhibitor has been approved yet. There is still demand to discover novel p38α to deal with the safety issue induced by off-target effects. OBJECTIVE: In this study, we performed a machine learning-based virtual screening to identify p38α inhibitors from a natural products library, expecting to find novel drug lead scaffolds. METHODS: Firstly, the training dataset was processed with similarity screening to fit the chemical space of the natural products library. Then, six classifiers were constructed by combing two sets of molecular features with three different machine learning algorithms. After model evaluation, the three best classifiers were used for virtual screening. RESULTS: Among the 15 compounds selected for experimental validation, picrasidine S was identified as a p38α inhibitor with the IC50 as 34.14 µM. Molecular docking was performed to predict the interaction mode of picrasidine S and p38α, indicating a specific hydrogen bond with Met109. CONCLUSION: This work provides a protocol and example for machine learning-assisted discovery of p38α inhibitor from natural products, as well as a novel lead scaffold represented by picrasidine S for further optimization and investigation.

kinCSM: Using graph-based signatures to predict small molecule CDK2 inhibitors.

Protein phosphorylation acts as an essential on/off switch in many cellular signaling pathways. This has led to ongoing interest in targeting kinases for therapeutic intervention. Computer-aided drug discovery has been proven a useful and cost-effective approach for facilitating prioritization and enrichment of screening libraries, but limited effort has been devoted providing insights on what makes a potent kinase inhibitor. To fill this gap, here we developed kinCSM, an integrative computational tool capable of accurately identifying potent cyclin-dependent kinase 2 (CDK2) inhibitors, quantitatively predicting CDK2 ligand-kinase inhibition constants (pKi ) and classifying different types of inhibitors based on their favorable binding modes. kinCSM predictive models were built using supervised learning and leveraged the concept of graph-based signatures to capture both physicochemical properties and geometry properties of small molecules. CDK2 inhibitors were accurately identified with Matthew's Correlation Coefficients (MCC) of up to 0.74, and inhibition constants predicted with Pearson's correlation of up to 0.76, both with consistent performances of 0.66 and 0.68 on a nonredundant blind test, respectively. kinCSM was also able to identify the potential type of inhibition for a given molecule, achieving MCC of up to 0.80 on cross-validation and 0.73 on the blind test. Analyzing the molecular composition of revealed enriched chemical fragments in CDK2 inhibitors and different types of inhibitors, which provides insights into the molecular mechanisms behind ligand-kinase interactions. kinCSM will be an invaluable tool to guide future kinase drug discovery. To aid the fast and accurate screening of CDK2 inhibitors, kinCSM is freely available at https://biosig.lab.uq.edu.au/kin_csm/.

An Ensemble Structure and Physicochemical (SPOC) Descriptor for Machine-Learning Prediction of Chemical Reaction and Molecular Properties.

Feature representations, or descriptors, are machines' chemical language that largely shapes the prediction capability, generalizability and interpretability of machine learning models. To develop a generally applicable descriptor is highly warranted for chemists to deal with conventional prediction tasks in the context of sparsely distributed and small datasets. Inspired by the chemist's vision on molecules, we presented herein an ensemble descriptor, SPOC, curated on the principles of physical organic chemistry that integrates Structure and Physicochemical property (SPOC) of a molecule. SPOC could be readily constructed by combining molecular fingerprints, representing the structure of a given molecule, and molecular physicochemical properties extracted from RDKit or Mordred molecular descriptors. The applicability of SPOC was fully surveyed in a range of well-structured chemical databases with machine learning tasks varying from regression to classifications.

MultiPep: a hierarchical deep learning approach for multi-label classification of peptide bioactivities.

Peptide-based therapeutics are here to stay and will prosper in the future. A key step in identifying novel peptide-drugs is the determination of their bioactivities. Recent advances in peptidomics screening approaches hold promise as a strategy for identifying novel drug targets. However, these screenings typically generate an immense number of peptides and tools for ranking these peptides prior to planning functional studies are warranted. Whereas a couple of tools in the literature predict multiple classes, these are constructed using multiple binary classifiers. We here aimed to use an innovative deep learning approach to generate an improved peptide bioactivity classifier with capacity of distinguishing between multiple classes. We present MultiPep: a deep learning multi-label classifier that assigns peptides to zero or more of 20 bioactivity classes. We train and test MultiPep on data from several publically available databases. The same data are used for a hierarchical clustering, whose dendrogram shapes the architecture of MultiPep. We test a new loss function that combines a customized version of Matthews correlation coefficient with binary cross entropy (BCE), and show that this is better than using class-weighted BCE as loss function. Further, we show that MultiPep surpasses state-of-the-art peptide bioactivity classifiers and that it predicts known and novel bioactivities of FDA-approved therapeutic peptides. In conclusion, we present innovative machine learning techniques used to produce a peptide prediction tool to aid peptide-based therapy development and hypothesis generation.

Identification of Broad-Spectrum MMP Inhibitors by Virtual Screening.

Matrix metalloproteinases (MMPs) are the family of proteases that are mainly responsible for degrading extracellular matrix (ECM) components. In the skin, the overexpression of MMPs as a result of ultraviolet radiation triggers an imbalance in the ECM turnover in a process called photoaging, which ultimately results in skin wrinkling and premature skin ageing. Therefore, the inhibition of different enzymes of the MMP family at a topical level could have positive implications for photoaging. Considering that the MMP catalytic region is mostly conserved across different enzymes of the MMP family, in this study we aimed to design a virtual screening (VS) workflow to identify broad-spectrum MMP inhibitors that can be used to delay the development of photoaging. Our in silico approach was validated in vitro with 20 VS hits from the Specs library that were not only structurally different from one another but also from known MMP inhibitors. In this bioactivity assay, 18 of the 20 compounds inhibit at least one of the assayed MMPs at 100 µM (with 5 of them showing around 50% inhibition in all the tested MMPs at this concentration). Finally, this VS was used to identify natural products that have the potential to act as broad-spectrum MMP inhibitors and be used as a treatment for photoaging.

Antimigratory effect of pyrazole derivatives through the induction of STAT1 phosphorylation in A549 cancer cells.

OBJECTIVES: In cancer treatment, it is important to prevent or slow down metastasis as well as preventing the proliferation of cancer cells. In this study, we aimed to find pyrazole compounds with antimigratory properties. METHODS: The 'PASSonline' programme was used to determine the possible pharmacological activities of the pyrazole compounds selected from the library, and two pyrazole derivatives were identified as a transcription factor STAT inhibitor with a high probability. There are studies known that JAK/STAT pathway is related to cancer cell migration, thus the possible antimigratory effects of these two synthesized pyrazole compounds were examined in A549 cancer cells. KEY FINDINGS: Our data demonstrated that compound-2 at different concentrations significantly inhibited cell migration in A549 cells. Then, the effects of these compounds on STAT activation were evaluated. We reported that 10 µM compound-2 induced a significant phosphorylation of STAT1 suggesting that STAT1 activation may be responsible for the antimigratory effect of compound-2. CONCLUSIONS: Taken together, the compound-2 is a promising compound with the antimigratory activity for cancer treatment, and further studies are needed to synthesize more active derivatives by evaluating the structure-activity relationship of leading compound-2.

Identification of the New In Vivo Metabolites of Ilaprazole in Rat Plasma after Oral Administration by LC-MS: In Silico Prediction of the H+/K+-ATPase Inhibitor.

Ilaprazole is a proton pump inhibitor used to treat digestive diseases. In this study, blood samples were collected after oral administration of ilaprazole and prepared by liquid-liquid extraction. The metabolites of ilaprazole were detected by liquid chromatography-high-resolution mass spectrometry (LC-HRMS) and LC-MSn. A total of twelve in vivo metabolites were detected in rat plasma and six new metabolites of ilaprazole, including one reductive metabolite with sulfide (M3), two hydroxylated metabolites with sulfoxide (M7 and M8), and three oxidative metabolites with sulfone (M9, M11, and M12), were identified. The possible metabolic pathways of ilaprazole and the fragmentation behaviors of its metabolites were elucidated. The result of the in silico prediction indicates that all the new metabolites showed the potential ability to inhibit H+/K+-ATPase activity.

Convolutional architectures for virtual screening.

BACKGROUND: A Virtual Screening algorithm has to adapt to the different stages of this process. Early screening needs to ensure that all bioactive compounds are ranked in the first positions despite of the number of false positives, while a second screening round is aimed at increasing the prediction accuracy. RESULTS: A novel CNN architecture is presented to this aim, which predicts bioactivity of candidate compounds on CDK1 using a combination of molecular fingerprints as their vector representation, and has been trained suitably to achieve good results as regards both enrichment factor and accuracy in different screening modes (98.55% accuracy in active-only selection, and 98.88% in high precision discrimination). CONCLUSION: The proposed architecture outperforms state-of-the-art ML approaches, and some interesting insights on molecular fingerprints are devised.

Bioactivity Prediction Based on Matched Molecular Pair and Matched Molecular Series Methods.

BACKGROUND: Enhancing a compound's biological activity is the central task for lead optimization in small molecules drug discovery. However, it is laborious to perform many iterative rounds of compound synthesis and bioactivity tests. To address the issue, it is highly demanding to develop high quality in silico bioactivity prediction approaches, to prioritize such more active compound derivatives and reduce the trial-and-error process. METHODS: Two kinds of bioactivity prediction models based on a large-scale structure-activity relationship (SAR) database were constructed. The first one is based on the similarity of substituents and realized by matched molecular pair analysis, including SA, SA_BR, SR, and SR_BR. The second one is based on SAR transferability and realized by matched molecular series analysis, including Single MMS pair, Full MMS series, and Multi single MMS pairs. Moreover, we also defined the application domain of models by using the distance-based threshold. RESULTS: Among seven individual models, Multi single MMS pairs bioactivity prediction model showed the best performance (R2 = 0.828, MAE = 0.406, RMSE = 0.591), and the baseline model (SA) produced the most lower prediction accuracy (R2 = 0.798, MAE = 0.446, RMSE = 0.637). The predictive accuracy could further be improved by consensus modeling (R2 = 0.842, MAE = 0.397 and RMSE = 0.563). CONCLUSION: An accurate prediction model for bioactivity was built with a consensus method, which was superior to all individual models. Our model should be a valuable tool for lead optimization.

SmartGraph: a network pharmacology investigation platform.

MOTIVATION: Drug discovery investigations need to incorporate network pharmacology concepts while navigating the complex landscape of drug-target and target-target interactions. This task requires solutions that integrate high-quality biomedical data, combined with analytic and predictive workflows as well as efficient visualization. SmartGraph is an innovative platform that utilizes state-of-the-art technologies such as a Neo4j graph-database, Angular web framework, RxJS asynchronous event library and D3 visualization to accomplish these goals. RESULTS: The SmartGraph framework integrates high quality bioactivity data and biological pathway information resulting in a knowledgebase comprised of 420,526 unique compound-target interactions defined between 271,098 unique compounds and 2018 targets. SmartGraph then performs bioactivity predictions based on the 63,783 Bemis-Murcko scaffolds extracted from these compounds. Through several use-cases, we illustrate the use of SmartGraph to generate hypotheses for elucidating mechanism-of-action, drug-repurposing and off-target prediction. AVAILABILITY: https://smartgraph.ncats.io/.

Pharmaceutical analysis combined with in-silico therapeutic and toxicological profiling on zileuton and its impurities to assist in modern drug discovery.

The obligatory testing of drug molecules and their impurities to protect users against toxic compounds seems to provide interesting opportunities for new drug discovery. Impurities, which proved to be non-toxic, may be explored for their own therapeutic potential and thus be a part of future drug discovery. The essential role of pharmaceutical analysis can thus be extended to achieve this purpose. The present study examined these objectives by characterizing the major degradation products of zileuton (ZLT), a 5-lipoxygenase (5-LOX) inhibitor being prevalently used to treat asthma. The drug sample was exposed to forced degradation and found susceptible to hydrolysis and oxidative stress. The obtained Forced Degradation Products (FDP's) were resolved using an earlier developed and validated Ultra-High-Pressure Liquid Chromatography Photo-Diode-Array (UHPLC-PDA) protocol. ZLT, along with acid-and alkali-stressed samples, were subjected to Liquid-chromatography Mass-spectrometry Quadrupole Time-of-flight (LC/MS-QTOF) studies. Major degradation products were isolated using Preparative TLC and characterized using Q-TOF and/or Proton nuclear magnetic resonance (1HNMR) studies. The information obtained was assembled for structural conformation. Toxicity Prediction using Komputer Assisted Technology (TOPKAT) toxicity analyses indicated some FDP's as non-toxic when compared to ZLT. Hence, these non-toxic impurities may have bio-affinity and can be explored to interact with other therapeutic targets, to assist in drug discovery. The drug molecule and the characterized FDP's were subjected to 3-Dimensional Extra Precision (3D-XP)-molecular docking to explore changes in bio-affinity for the 5-LOX enzyme (PDB Id: 3V99). One FDP was found to have a higher binding affinity than the drug itself, indicating it may be a suitable antiasthmatic. The possibility of being active at other sites cannot be neglected and this is evaluated to a reasonable extent by Prediction of Activity Spectra for Substances (PASS). Besides being antiasthmatic, some FDP's were predicted antineoplastic, antiallergic and inhibitors of Complement Factor-D.

In-Silico Molecular Binding Prediction for Human Drug Targets Using Deep Neural Multi-Task Learning.

In in-silico prediction for molecular binding of human genomes, promising results have been demonstrated by deep neural multi-task learning due to its strength in training tasks with imbalanced data and its ability to avoid over-fitting. Although the interrelation between tasks is known to be important for successful multi-task learning, its adverse effect has been underestimated. In this study, we used molecular interaction data of human targets from ChEMBL to train and test various multi-task and single-task networks and examined the effectiveness of multi-task learning for different compositions of targets. Targets were clustered based on sequence similarity in their binding domains and various target sets from clusters were chosen. By comparing the performance of deep neural architectures for each target set, we found that similarity within a target set is highly important for reliable multi-task learning. For a diverse target set or overall human targets, the performance of multi-task learning was lower than single-task learning, but outperformed single-task for the target set containing similar targets. From this insight, we developed Multiple Partial Multi-Task learning, which is suitable for binding prediction for human drug targets.

Elucidating Compound Mechanism of Action and Predicting Cytotoxicity Using Machine Learning Approaches, Taking Prediction Confidence into Account.

The modes of action (MoAs) of drugs frequently are unknown, because many are small molecules initially identified from phenotypic screens, giving rise to the need to elucidate their MoAs. In addition, the high attrition rate for candidate drugs in preclinical studies due to intolerable toxicity has motivated the development of computational approaches to predict drug candidate (cyto)toxicity as early as possible in the drug-discovery process. Here, we provide detailed instructions for capitalizing on bioactivity predictions to elucidate the MoAs of small molecules and infer their underlying phenotypic effects. We illustrate how these predictions can be used to infer the underlying antidepressive effects of marketed drugs. We also provide the necessary functionalities to model cytotoxicity data using single and ensemble machine-learning algorithms. Finally, we give detailed instructions on how to calculate confidence intervals for individual predictions using the conformal prediction framework. © 2019 by John Wiley & Sons, Inc.

The Light and Dark Sides of Virtual Screening: What Is There to Know?

Virtual screening consists of using computational tools to predict potentially bioactive compounds from files containing large libraries of small molecules. Virtual screening is becoming increasingly popular in the field of drug discovery as in silico techniques are continuously being developed, improved, and made available. As most of these techniques are easy to use, both private and public organizations apply virtual screening methodologies to save resources in the laboratory. However, it is often the case that the techniques implemented in virtual screening workflows are restricted to those that the research team knows. Moreover, although the software is often easy to use, each methodology has a series of drawbacks that should be avoided so that false results or artifacts are not produced. Here, we review the most common methodologies used in virtual screening workflows in order to both introduce the inexperienced researcher to new methodologies and advise the experienced researcher on how to prevent common mistakes and the improper usage of virtual screening methodologies.