Targeting Hypoglycemic Natural Products from the Cloud Forest Plants Using Chemotaxonomic Computer-Assisted Selection.

The cloud forest (CF), a hugely biodiverse ecosystem, is a hotspot of unexplored plants with potential for discovering pharmacologically active compounds. Without sufficient ethnopharmacological information, developing strategies for rationally selecting plants for experimental studies is crucial. With this goal, a CF metabolites library was created, and a ligand-based virtual screening was conducted to identify molecules with potential hypoglycemic activity. From the most promising botanical families, plants were collected, methanolic extracts were prepared, and hypoglycemic activity was evaluated through in vitro enzyme inhibition assays on α-amylase, α-glucosidase, and dipeptidyl peptidase IV (DPP-IV). Metabolomic analyses were performed to identify the dominant metabolites in the species with the best inhibitory activity profile, and their affinity for the molecular targets was evaluated using ensemble molecular docking. This strategy led to the identification of twelve plants (in four botanical families) with hypoglycemic activity. Sida rhombifolia (Malvaceae) stood out for its DPP-IV selective inhibition versus S. glabra. A comparison of chemical profiles led to the annotation of twenty-seven metabolites over-accumulated in S. rhombifolia compared to S. glabra, among which acanthoside D and cis-tiliroside were noteworthy for their potential selective inhibition due to their specific intermolecular interactions with relevant amino acids of DPP-IV. The workflow used in this study presents a novel targeting strategy for identifying novel bioactive natural sources, which can complement the conventional selection criteria used in Natural Product Chemistry.

Conformational Space Profiling Enhances Generic Molecular Representation for AI-Powered Ligand-Based Drug Discovery.

The molecular representation model is a neural network that converts molecular representations (SMILES, Graph) into feature vectors, and is an essential module applied across a wide range of artificial intelligence-driven drug discovery scenarios. However, current molecular representation models rarely consider the three-dimensional conformational space of molecules, losing sight of the dynamic nature of small molecules as well as the essence of molecular conformational space that covers the heterogeneity of molecule properties, such as the multi-target mechanism of action, recognition of different biomolecules, dynamics in cytoplasm and membrane. In this study, a new model named GeminiMol is proposed to incorporate conformational space profiles into molecular representation learning, which extracts the feature of capturing the complicated interplay between the molecular structure and the conformational space. Although GeminiMol is pre-trained on a relatively small-scale molecular dataset (39290 molecules), it shows balanced and superior performance not only on 67 molecular properties predictions but also on 73 cellular activity predictions and 171 zero-shot tasks (including virtual screening and target identification). By capturing the molecular conformational space profile, the strategy paves the way for rapid exploration of chemical space and facilitates changing paradigms for drug design.

Targeting COVID-19 and varicocele by blocking inflammasome: Ligand-based virtual screening.

COVID-19 is a new generation of outbreaks that invade not only local emerging region, continental but also the whole globe. Varicocele on the other hand, is a testicular vascular disease that underlies 40 % of male infertility cases. Fortunately, the two diseases can be blocked through targeting one common target, NLRP3 inflammasome. Upon searching for similar drugs that gained FDA-approval in ChEMBL library along with examining their potential blockade of the receptor through docking using CB-DOCK-2, three potential approved drugs can be repurposed, ChEMBL 4297185, ChEMBL 1201749, ChEMBL 1200545 which had binding energy of -9.8 and -9.7 kcal/mol (stronger than the reference inhibitor, -9.3 kcal/mol). Also, ADME profile of the top 3 drugs showed better attributes. Also, the simulated proteins exhibited stable pattern with strong free binding energies. Among the potential inhibitor drugs ChEMBL 4297185 was found to remain inside the binding site of the protein during the 200 ns simulation time. Hence, it is anticipated to have the highest binding and thus inhibition potential against the protein. The suggested drugs, especially ChEMBL 4297185, are potentially repurposable toward treating COVID-19 and varicocele which deserve further experimental validation.

Identifying and characterising promising small molecule inhibitors of kinesin spindle protein using ligand-based virtual screening, molecular docking, molecular dynamics and MM­GBSA calculations.

The kinesin spindle protein (Eg5) is a mitotic protein that plays an essential role in the formation of the bipolar spindles during the mitotic phase. Eg5 protein controls the segregation of the chromosomes in mitosis which renders it a vital target for cancer treatment. In this study our approach to identifying novel scaffold for Eg5 inhibitors is based on targeting the novel allosteric pocket (α4/α6/L11). Extensive computational techniques were applied using ligand-based virtual screening and molecular docking by two approaches, MOE and AutoDock, to screen a library of commercial compounds. We identified compound 8-(3-(1H-imidazol-1-ylpropylamino)-3-methyl-7-((naphthalen-3-yl)methyl)-1H-purine-2, 6 (3H,7H)-dione (compound 5) as a novel scaffold for Eg5 inhibitors. This compound inhibited cancer cell Eg5 ATPase at 2.37 ± 0.15 µM. The molecular dynamics simulations revealed that the identified compound formed stable interactions in the allosteric pocket (α4/α6/L11) of the receptor, indicating its potential as a novel Eg5 inhibitor.

A Combination of Pharmacophore Generation, Ligand-based Virtual Screening, Atom-based 3D-QSAR, and Molecular Docking Studies on Febuxostat-based Amides Analogues as Anti-inflammatory Agents.

BACKGROUND: A defence mechanism of the body includes inflammation. It is a process through which the immune system identifies, rejects, and starts to repair foreign and damaging stimuli. In the world, chronic inflammatory disorders are the leading cause of death. MATERIAL AND METHODS: To obtain optimized pharmacophore, previously reported febuxostat- based anti-inflammatory amide derivatives series were subjected to pharmacophore hypothesis, ligand-based virtual screening, and 3D-QSAR studies in the present work using Schrodinger suite 2022-4. QuikProp module of Schrodinger was used for ADMET prediction, and HTVS, SP, and XP protocols of GLIDE modules were used for molecular docking on target protein (PDB ID:3LN1). RESULT: Utilising 29 compounds, a five-point model of common pharmacophore hypotheses was created, having pIC50 ranging between 5.34 and 4.871. The top pharmacophore hypothesis AHHRR_ 1 model consists of one hydrogen bond acceptor, two hydrophobic groups and two ring substitution features. The hypothesis model AHHRR_1 underwent ligand-based virtual screening using the molecules from Asinex. Additionally, a 3D-QSAR study based on individual atoms was performed to assess their contributions to model development. The top QSAR model was chosen based on the values of R2 (0.9531) and Q2 (0.9424). Finally, four potential hits were obtained by molecular docking based on virtual screening. CONCLUSION: The virtual screen compounds have shown similar docking interaction with amino acid residues as shown by standard diclofenac sodium drugs. Therefore, the findings in the present study can be explored in the development of potent anti-inflammatory agents.

SpaceGrow: efficient shape-based virtual screening of billion-sized combinatorial fragment spaces.

The growing size of make-on-demand chemical libraries is posing new challenges to cheminformatics. These ultra-large chemical libraries became too large for exhaustive enumeration. Using a combinatorial approach instead, the resource requirement scales approximately with the number of synthons instead of the number of molecules. This gives access to billions or trillions of compounds as so-called chemical spaces with moderate hardware and in a reasonable time frame. While extremely performant ligand-based 2D methods exist in this context, 3D methods still largely rely on exhaustive enumeration and therefore fail to apply. Here, we present SpaceGrow: a novel shape-based 3D approach for ligand-based virtual screening of billions of compounds within hours on a single CPU. Compared to a conventional superposition tool, SpaceGrow shows comparable pose reproduction capacity based on RMSD and superior ranking performance while being orders of magnitude faster. Result assessment of two differently sized subsets of the eXplore space reveals a higher probability of finding superior results in larger spaces highlighting the potential of searching in ultra-large spaces. Furthermore, the application of SpaceGrow in a drug discovery workflow was investigated in four examples involving G protein-coupled receptors (GPCRs) with the aim to identify compounds with similar binding capabilities and molecular novelty.

The role and potential of computer-aided drug discovery strategies in the discovery of novel antimicrobials.

Antimicrobial resistance (AMR) has become more of a concern in recent decades, particularly in infections associated with global public health threats. The development of new antibiotics is crucial to ensuring infection control and eradicating AMR. Although drug discovery and development are essential processes in the transformation of a drug candidate from the laboratory to the bedside, they are often very complicated, expensive, and time-consuming. The pharmaceutical sector is continuously innovating strategies to reduce research costs and accelerate the development of new drug candidates. Computer-aided drug discovery (CADD) has emerged as a powerful and promising technology that renews the hope of researchers for the faster identification, design, and development of cheaper, less resource-intensive, and more efficient drug candidates. In this review, we discuss an overview of AMR, the potential, and limitations of CADD in AMR drug discovery, and case studies of the successful application of this technique in the rapid identification of various drug candidates. This review will aid in achieving a better understanding of available CADD techniques in the discovery of novel drug candidates against resistant pathogens and other infectious agents.

Decrypting orphan GPCR drug discovery via multitask learning.

The drug discovery of G protein-coupled receptors (GPCRs) superfamily using computational models is often limited by the availability of protein three-dimensional (3D) structures and chemicals with experimentally measured bioactivities. Orphan GPCRs without known ligands further complicate the process. To enable drug discovery for human orphan GPCRs, multitask models were proposed for predicting half maximal effective concentrations (EC50) of the pairs of chemicals and GPCRs. Protein multiple sequence alignment features, and physicochemical properties and fingerprints of chemicals were utilized to encode the protein and chemical information, respectively. The protein features enabled the transfer of data-rich GPCRs to orphan receptors and the transferability based on the similarity of protein features. The final model was trained using both agonist and antagonist data from 200 GPCRs and showed an excellent mean squared error (MSE) of 0.24 in the validation dataset. An independent test using the orphan dataset consisting of 16 receptors associated with less than 8 bioactivities showed a reasonably good MSE of 1.51 that can be further improved to 0.53 by considering the transferability based on protein features. The informative features were identified and mapped to corresponding 3D structures to gain insights into the mechanism of GPCR-ligand interactions across the GPCR family. The proposed method provides a novel perspective on learning ligand bioactivity within the diverse human GPCR superfamily and can potentially accelerate the discovery of therapeutic agents for orphan GPCRs.

Identification of new small molecule allosteric SHP2 inhibitor through pharmacophore-based virtual screening, molecular docking, molecular dynamics simulation studies, synthesis and in vitro evaluation.

Src homology-2 (SH2) domain-containing phosphatase-2 (SHP2) is the first identified protooncogene and is a promising target for developing small molecule inhibitors as cancer chemotherapeutic agents. Pharmacophore-based virtual screening (PBVS) is a pharmacoinformatics methodology that employs physicochemical knowhow of the chemical space into the dynamic environs of computational technology to extract virtual molecular hits that are precise and promising for a drug target. In the current study, PBVS has been applied on EnamineTM Advanced Collection of 551,907 molecules by using a pharmacophore model developed upon SHP099 by Molecular Operating Environment (MOE) software to identify potential small molecule allosteric SHP2 inhibitors. Obtained 37 hits were further filtered through DruLiTo software for drug-likeness and PAINS remover which yielded 35 hits. These were subjected to molecular docking studies against the tunnel allosteric site of SHP2 (PDB ID: 5EHR) to screen them according to their binding affinity for the enzyme. Top 5 molecules having highest binding affinity for 5EHR were passed through an ADMET prediction screening and the top 2 hits (ligands 111675 and 546656) with the most favourable ADMET profile were taken for post screening molecular docking and MD simulation studies. From the protein-ligand interaction pattern, conformational stability and energy parameters, ligand 111675 (SHP2 Ki = 0.118 µM) resulted as the most active molecule. Further, the synthesis and in vitro evaluation of the lead compound 111675 unveiled its potent inhibitory activity (IC50 = 0.878 ± 0.008 µM) against SHP2.Communicated by Ramaswamy H. Sarma.

Computational Screening Using a Combination of Ligand-Based Machine Learning and Molecular Docking Methods for the Repurposing of Antivirals Targeting the SARS-CoV-2 Main Protease.

BACKGROUND: COVID-19 is an infectious disease caused by SARS-CoV-2, a close relative of SARS-CoV. Several studies have searched for COVID-19 therapies. The topics of these works ranged from vaccine discovery to natural products targeting the SARS-CoV-2 main protease (Mpro), a potential therapeutic target due to its essential role in replication and conserved sequences. However, published research on this target is limited, presenting an opportunity for drug discovery and development. METHOD: This study aims to repurpose 10692 drugs in DrugBank by using ligand-based virtual screening (LBVS) machine learning (ML) with Konstanz Information Miner (KNIME) to seek potential therapeutics based on Mpro inhibitors. The top candidate compounds, the native ligand (GC-376) of the Mpro inhibitor, and the positive control boceprevir were then subjected to absorption, distribution, metabolism, excretion, and toxicity (ADMET) characterization, drug-likeness prediction, and molecular docking (MD). Protein-protein interaction (PPI) network analysis was added to provide accurate information about the Mpro regulatory network. RESULTS: This study identified 3,166 compound candidates inhibiting Mpro. The random forest (RF) molecular access system ML model provided the highest confidence score of 0.95 (bromo-7-nitroindazole) and identified the top 22 candidate compounds. Subjecting the 22 candidate compounds, the native ligand GC-376, and boceprevir to further ADMET property characterization and drug-likeness predictions revealed that one compound had two violations of Lipinski's rule. Additional MD results showed that only five compounds had more negative binding energies than the native ligand (- 12.25 kcal/mol). Among these compounds, CCX-140 exhibited the lowest score of - 13.64 kcal/mol. Through literature analysis, six compound classes with potential activity for Mpro were discovered. They included benzopyrazole, azole, pyrazolopyrimidine, carboxylic acids and derivatives, benzene and substituted derivatives, and diazine. Four pathologies were also discovered on the basis of the Mpro PPI network. CONCLUSION: Results demonstrated the efficiency of LBVS combined with MD. This combined strategy provided positive evidence showing that the top screened drugs, including CCX-140, which had the lowest MD score, can be reasonably advanced to the in vitro phase. This combined method may accelerate the discovery of therapies for novel or orphan diseases from existing drugs.

Repurposing Drugs for Inhibition against ALDH2 via a 2D/3D Ligand-Based Similarity Search and Molecular Simulation.

Aldehyde dehydrogenase-2 (ALDH2) is a crucial enzyme participating in intracellular aldehyde metabolism and is acknowledged as a potential therapeutic target for the treatment of alcohol use disorder and other addictive behaviors. Using previously reported ALDH2 inhibitors of Daidzin, CVT-10216, and CHEMBL114083 as reference molecules, here we perform a ligand-based virtual screening of world-approved drugs via 2D/3D similarity search methods, followed by the assessments of molecular docking, toxicity prediction, molecular simulation, and the molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) analysis. The 2D molecular fingerprinting of ECFP4 and FCFP4 and 3D molecule-shape-based USRCAT methods show good performances in selecting compounds with a strong binding behavior with ALDH2. Three compounds of Zeaxanthin (q = 0), Troglitazone (q = 0), and Sequinavir (q = +1 e) are singled out as potential inhibitors; Zeaxanthin can only be hit via USRCAT. These drugs displayed a stronger binding strength compared to the reported potent inhibitor CVT-10216. Sarizotan (q = +1 e) and Netarsudil (q = 0/+1 e) displayed a strong binding strength with ALDH2 as well, whereas they displayed a shallow penetration into the substrate-binding tunnel of ALDH2 and could not fully occupy it. This likely left a space for substrate binding, and thus they were not ideal inhibitors. The MM-PBSA results indicate that the selected negatively charged compounds from the similarity search and Vina scoring are thermodynamically unfavorable, mainly due to electrostatic repulsion with the receptor (q = -6 e for ALDH2). The electrostatic attraction with positively charged compounds, however, yielded very strong binding results with ALDH2. These findings reveal a deficiency in the modeling of electrostatic interactions (in particular, between charged moieties) in the virtual screening via the 2D/3D similarity search and molecular docking with the Vina scoring system.

Mechanisms of Nelumbinis folium targeting PPARγ for weight management: A molecular docking and molecular dynamics simulations study.

The lotus leaf, Nelumbinis folium (NF), has frequently appeared in obesity clinical trials as an intervention to promote weight loss and improve metabolic profiles. However, the molecular mechanisms by which it interacts with important obesity targets and pathways, such as the peroxisome proliferator-activated receptor gamma (PPARγ) within the PPAR signalling pathway, were not well understood. This study aims to screen for candidate compounds from NF with desirable pharmacokinetic properties and examine their binding feasibility at the PPARγ ligand-binding domain (LBD). Ligand- and structure-based screening of NF compounds were performed, and a consensus approach has been applied to identify druggable candidates. By examining the pharmacokinetic profiles, a large proportion of NF compounds exhibited favourable drug-likeness and oral bioavailability properties. Furthermore, the binding affinity scores and poses provided new insights on the distinctive binding behaviours of NF compounds at the LBD of PPARγ in its inactive form. Several NF compounds could bind strongly to PPARγ at sub-pockets where partial agonists and antagonists were found to bind and may induce conformational changes that influence co-repressor binding, trans-repression, and gene expression inhibition. Subsequent molecular dynamics simulations of a candidate compound (NF129 narcissin) bound to PPARγ revealed conformational stability, residue fluctuation, and binding behaviours comparable to that of the known inhibitor, SR1664. Therefore, it can be proposed that narcissin exhibits characteristics of a PPARγ antagonist. Further experimental validation to support the development of NF129 as a future anti-obesity agent is warranted.

An Optimized Cloud Computing Method for Extracting Molecular Descriptors.

Extracting molecular descriptors from chemical compounds is an essential preprocessing phase for developing accurate classification models. Supervised machine learning algorithms offer the capability to detect "hidden" patterns that may exist in a large dataset of compounds, which are represented by their molecular descriptors. Assuming that molecules with similar structure tend to share similar physicochemical properties, large chemical libraries can be screened by applying similarity sourcing techniques in order to detect potential bioactive compounds against a molecular target. However, the process of generating these compound features is time-consuming. Our proposed methodology not only employs cloud computing to accelerate the process of extracting molecular descriptors but also introduces an optimized approach to utilize the computational resources in the most efficient way.

Ligand-Based Virtual Screening, Molecular Docking, and Molecular Dynamic Simulations of New ß-Estrogen Receptor Activators with Potential for Pharmacological Obesity Treatment.

Obesity is a pandemic and a serious health problem in developed and undeveloped countries. Activation of estrogen receptor beta (ERß) has been shown to promote weight loss without modifying caloric intake, making it an attractive target for developing new drugs against obesity. This work aimed to predict new small molecules as potential ERß activators. A ligand-based virtual screening of the ZINC15, PubChem, and Molport databases by substructure and similarity was carried out using the three-dimensional organization of known ligands as a reference. A molecular docking screening of FDA-approved drugs was also conducted as a repositioning strategy. Finally, selected compounds were evaluated by molecular dynamic simulations. Compounds 1 (-24.27 ± 0.34 kcal/mol), 2 (-23.33 ± 0.3 kcal/mol), and 6 (-29.55 ± 0.51 kcal/mol) showed the best stability on the active site in complex with ERß with an RMSD < 3.3 Å. RMSF analysis showed that these compounds do not affect the fluctuation of the Cα of ERß nor the compactness according to the radius of gyration. Finally, an in silico evaluation of ADMET showed they are safe molecules. These results suggest that new ERß ligands could be promising molecules for obesity control.

Virtual and In Vitro Screening of Natural Products Identifies Indole and Benzene Derivatives as Inhibitors of SARS-CoV-2 Main Protease (Mpro).

The rapid spread of the coronavirus disease 2019 (COVID-19) resulted in serious health, social, and economic consequences. While the development of effective vaccines substantially reduced the severity of symptoms and the associated deaths, we still urgently need effective drugs to further reduce the number of casualties associated with SARS-CoV-2 infections. Machine learning methods both improved and sped up all the different stages of the drug discovery processes by performing complex analyses with enormous datasets. Natural products (NPs) have been used for treating diseases and infections for thousands of years and represent a valuable resource for drug discovery when combined with the current computation advancements. Here, a dataset of 406,747 unique NPs was screened against the SARS-CoV-2 main protease (Mpro) crystal structure (6lu7) using a combination of ligand- and structural-based virtual screening. Based on 1) the predicted binding affinities of the NPs to the Mpro, 2) the types and number of interactions with the Mpro amino acids that are critical for its function, and 3) the desirable pharmacokinetic properties of the NPs, we identified the top 20 candidates that could potentially inhibit the Mpro protease function. A total of 7 of the 20 top candidates were subjected to in vitro protease inhibition assay and 4 of them (4/7; 57%), including two beta carbolines, one N-alkyl indole, and one Benzoic acid ester, had significant inhibitory activity against Mpro protease. These four NPs could be developed further for the treatment of COVID-19 symptoms.

In silico design, synthesis and evaluation of a less toxic octinoxate alternative with suitable photoprotection properties.

Excessive UV exposure leads to several skin pathologies such as sunburns, photoaging and carcinogenesis. Currently, sunscreen use is the most important factor in protecting skin from photoinduced damage. Octinoxate is a commonly used UV filter, but its use has become controversial because it acts as an endocrine disruptor in both humans, and marine animals. Research has relied on biotechnology, structure activity relationship (SAR) studies and combinatorial chemistry to find new and less toxic UV filters. However, there are no current examples that describe the possible applications of in silico techniques for obtaining these compounds. Thus, this project sought to design an octinoxate analog that could be used as a less toxic, but equally effective, photoprotective alternative through ligand based virtual screening (LBVS). We designed 213 novel molecules based on the (E)-cinnamoyl moiety of octinoxate, but only 23 were found to be less toxic than the parent compound. Then, an artificial neural network (ANN) based model was built to predict the molar absorptivity of those 23 molecules, and the molecule that presented a similar molar absorptivity to that of octinoxate was chosen for synthesis (analog 4, 3-phenylpropyl (E)-3-(4-methoxyphenyl)acrylate). Synthesis for analog 4 resulted in a 90% yield, and its photoprotective properties, lipophilicity and cytotoxicity were then evaluated. Analog 4 absorbed UV radiation in the range of 250-340 nm, and it presented a molar absorptivity of 36,155 M - 1cm-1. Its lipophilicity was evaluated with RP-HPLC resulting in a logkw of 2.49 and its LC50 was greater than octinoxate's (67.41 nM vs. 45.67 nM). Therefore, results showed that ligand based virtual screening is an effective strategy for the development of new organic UV filters, because it guided the design of less toxic analogs and pinpointed the most likely analog to exhibit UV properties similar to those of octinoxate. In this case, analog 4 is a promising alternative to its parent compound since it proved to be more effective and less toxic.

Identification of new small molecule monoamine oxidase-B inhibitors through pharmacophore-based virtual screening, molecular docking and molecular dynamics simulation studies.

The discovery of a safe and efficacious drug is a complex, time-consuming, and expensive process. Computational methodologies driven by cheminformatics tools play a central role in the high-throughput lead discovery and optimization process especially when the structure of the biological target is known. Monoamine oxidases are the membrane-bound FAD-containing enzymes and the isoform monoamine oxidase-B (MAO-B) is an attractive target for treating diseases like Alzheimer's disease, Parkinson's disease, glioma, etc. In the current study, we have used a pharmacophore-based virtual screening technique for the identification of new small molecule MAO-B inhibitors. Safinamide was used for building a pharmacophore model and the developed model was used to probe the ZINC database for potential hits. The obtained hits were filtered against drug-likeness and PAINS. Out of the hit's library, two compounds ZINC02181408, ZINC08853942 (most active), and ZINC53327382 (least active) were further subjected to molecular docking and dynamics simulation studies to assess their virtual binding affinities and stability of the resultant protein-ligand complex. The docking studies revealed that active ligands were well accommodated within the active site of MAO-B and interacted with both substrate and entrance cavity residues. MD simulation studies unveiled additional hydrogen bond interactions with the substrate cavity residues, Tyr398 and Tyr435 that are crucial for the catalytic role of MAO-B. Moreover, the predicted ADMET parameters suggest that the compounds ZINC08853942 and ZINC02181408 are suitable for CNS penetration. Thus, the attempted computational campaign yielded two potential MAO-B inhibitors that merit further experimental investigation.Communicated by Ramaswamy H. Sarma.

On the ability of machine learning methods to discover novel scaffolds.

The recent advances in the application of machine learning to drug discovery have made it a 'hot topic' for research, with hundreds of academic groups and companies integrating machine learning into their drug discovery projects. Nevertheless, there remains great uncertainty regarding the most appropriate ways to evaluate the relative performance of these powerful methods against more traditional cheminformatics approaches, and many pitfalls remain for the unwary. In 2020, researchers at MIT (Stokes et al., Cell 180(4), 688-702, 2020) reported the discovery of a new compound with antibacterial activity, halicin, through the use of a neural network machine learning method. A robust ability to identify new active chemotypes through computational methods would be very useful. In this study, we have used the Stokes et al. dataset to compare the performance of this method to two other approaches, Mapping of Activity Through Dichotomic Scores (MADS) by Todeschini et al. (J Chemom 32(4):e2994, 2018) and Random Matrix Theory (RMT) by Lee et al. (Proc Natl Acad Sci 116(9):3373-3378, 2019). Our results demonstrate that all three methods are capable of predicting halicin as an active antibacterial compound, but that this result is dependent on the dataset composition, pre-processing and the molecular fingerprint used. We have further assessed overall performance as determined by several performance metrics. We also investigated the scaffold hopping potential of the methods by modifying the dataset by removal of the ß-lactam and fluoroquinolone chemotypes. MADS and RMT are able to identify actives in the test set that contained these substructures. This ability arises because of high scoring fragments of the withheld chemotypes that are in common with other active antibiotic classes. Interestingly, MADS is relatively better compared to the other two methods based on general predictive performance.

Review on Compounds Isolated from Eriocaulaceae Family and Evaluation of Biological Activities by Machine Learning.

Eriocaulaceae is a pantropical family whose main center of biodiversity is in Brazil. In general, the family has about 1200 species, in which phytochemical and biological studies have shown a variety of structures and activities. The aim of this research is to compile the compounds isolated in the Eriocaulaceae family and carry out a computational study on their biological targets. The bibliographic research was carried out on six databases. Tables were built and organized according to the chemical class. In addition, a summary of the methods of isolating the compounds was also made. In the computational study were used ChEMBL platform, DRAGON 7.0, and the KNIME 4.4.0 software. Two hundred and twenty-two different compounds have been isolated in sixty-eight species, divided mainly into flavonoids and naphthopyranones, and minor compounds. The ligand-based virtual screening found promising molecules and molecules with multitarget potential, such as xanthones 194, 196, 200 and saponin 202, with xanthone 194 as the most promising. Several compounds with biological activities were isolated in the family, but the chemical profiles of many species are still unknown. The selected structures are a starting point for further studies to develop new antiparasitic and antiviral compounds based on natural products.

In-Silico Drug Toxicity and Interaction Prediction for Plant Complexes Based on Virtual Screening and Text Mining.

Potential drug toxicities and drug interactions of redundant compounds of plant complexes may cause unexpected clinical responses or even severe adverse events. On the other hand, super-additivity of drug interactions between natural products and synthetic drugs may be utilized to gain better performance in disease management. Although without enough datasets for prediction model training, based on the SwissSimilarity and PubChem platforms, for the first time, a feasible workflow of prediction of both toxicity and drug interaction of plant complexes was built in this study. The optimal similarity score threshold for toxicity prediction of this system is 0.6171, based on an analysis of 20 different herbal medicines. From the PubChem database, 31 different sections of toxicity information such as "Acute Effects", "NIOSH Toxicity Data", "Interactions", "Hepatotoxicity", "Carcinogenicity", "Symptoms", and "Human Toxicity Values" sections have been retrieved, with dozens of active compounds predicted to exert potential toxicities. In Spatholobus suberectus Dunn (SSD), there are 9 out of 24 active compounds predicted to play synergistic effects on cancer management with various drugs or factors. The synergism between SSD, luteolin and docetaxel in the management of triple-negative breast cancer was proved by the combination index assay, synergy score detection assay, and xenograft model.