Prediction Of Biological Activity Research Articles

Representative feature selection of molecular descriptors in QSAR modeling

Quantitative structure-activity relationship (QSAR) has been widely applied to many fields such as molecular toxicity detection and biological activity predictions. The screening of molecular descriptors (MDs) is a preliminary stage for building the QSAR model. An appropriate MDs set significantly impacts on the performance of QSAR model. Traditional screening of MDs is usually executed by artificial selection strongly depending on expertise and experience, which is difficult to cover all valuable cross-information among these dispersed and weak chemical information representative MDs. In this paper, we proposed a Representative Feature Selection (RFS) method that is capable of forming a representative set of MDs for QSAR model by calculating the Euclidean distances and Pearson correlation coefficients. The experimental result revealed that RFS effectively selects representative features from the feature space with information redundancy, and RFS enhances the performances of QSAR model.

Insilico Prediction of Biological Activity of Secondary Metabolites of Amphora sp.: for Development of Natural Medicines

Diatom Amphora sp. is a type of aquaculture product that has various types of metabolites with very good potential to be used as basic ingredients for drugs or other health products. Insilico analysis of extracts of Amphora sp. was carried out to predict biological activity that can be used as a basis for developing the use of aquaculture products. The research was conducted by extracting Amphora sp. using ethanol solvent with sonication method, the extraction results were analyzed by LC-HRMS, and the metabolite readings were then analyzed for drug potential and biological activity online using swissADME, Protox II, and Way2Drug applications. The results of the LC-HRMS analysis showed that the extracted Amphora sp. showed the highest abundance of valine compounds with 17.6% as well as other compounds L-Norleucin, stearamide, palmitoleic acid, isotretinoin, and arachidonic acid. The results of the ADME/T analysis showed that the six compounds of Amphora sp. extract had quite good potential according to the Lipinski rule of five. Biological activity also shows potential as an anti-inflammatory, antifungal, antibacterial, antiviral, and antineoplastic agent with Pa values ranging from 0.260 to 0.873. These results indicate that the production development of Amphora sp. and its use is very wide, including in the health sector. Keyword: ADME/T, PASSonline, Amphora sp., sonication, aquaculture

Chromenol Derivatives as Novel Antifungal Agents: Synthesis, In Silico and In Vitro Evaluation.

Herein we report the synthesis of some new 1H-1,2,4-triazole functionalized chromenols (3a–3n) via tandem reactions of 1-(alkyl/aryl)-2-(1H-1,2,4-triazole-1-yl) with salicylic aldehydes and the evaluation of their antifungal activity. In silico prediction of biological activity with computer program PASS indicate that the compounds have a high novelty compared to the known antifungal agents. We did not find any close analog among the over 580,000 pharmaceutical agents in the Cortellis Drug Discovery Intelligence database at the similarity cutoff of 70%. The evaluation of antifungal activity in vitro revealed that the highest activity was exhibited by compound 3k, followed by 3n. Their MIC values for different fungi were 22.1–184.2 and 71.3–199.8 µM, respectively. Twelve from fourteen tested compounds were more active than the reference drugs ketoconazole and bifonazole. The most sensitive fungus appeared to be Trichoderma viride, while Aspergillus fumigatus was the most resistant one. It was found that the presence of the 2-(tert-butyl)-2H-chromen-2-ol substituent on the 4th position of the triazole ring is very beneficial for antifungal activity. Molecular docking studies on C. albicans sterol 14α-demethylase (CYP51) and DNA topoisomerase IV were used to predict the mechanism of antifungal activities. According to the docking results, the inhibition of CYP51 is a putative mechanism of antifungal activity of the novel chromenol derivatives. We also showed that most active compounds have a low cytotoxicity, which allows us to consider them promising antifungal agents for the subsequent testing activity in in vivo assays.

Abstract 278: Artificial intelligence accelerate drug discovery

Abstract Drug discovery is resource intensive, and involves typical timelines of 10-20 years and costs that range from US$0.5 billion to US$2.6 billion. Artificial intelligence (AI) has recently started to gear-up its application in various sectors of the society and the pharmaceutical industry as a frontrunner beneficiary.Artificial intelligence can accelerate drug discovery and reduce costs by facilitating the rapid screening and identification of compounds. We have developed DM-AI drug discovery platform, including convolutional neural networks, decision treealgorithm, reinforcement learning, generative adversarial networks, big data, and knowledge graphs, along with structure and ligand-based high-throughput virtual screening , for new drug discovery and development. DM-AI optimizes biological activity,toxicity,physicochemical property. We used DM-AI to discover potent inhibitors of SHP2, PIM1, DNA-PK, kinases target implicated in solid tumor and other diseases.We started to train a biological activity prediction model on a database of the given target kinase inhibitors (positive set) and non-kinase targets molecules (negative set), and then predicted the activity of existing million data sets, obtained an initial output of thousands structures. We then evaluated these structures using a pharmacophore reward model on the basis of virtual chemical spaces of kinase inhibitors in complex with target protein. To narrow our focus to a smaller set of molecules for analysis, compounds with higher score were filtered to remove patents and applications molecules, also remove molecules bearing structural alerts and reactive groups.By day 7 after target selection, We had selected dozens structures with structural diversity for experimental validation. and by day 28, they were tested for in vitro inhibitory activity in an enzymatic kinase assay, active compounds accounted for up to 65% in some target models. This illustrates the utility of our DM-AI drug discovery platform for the successful, rapid discovery of drug candidates. Citation Format: Weidong Xie, Xing Cheng, Zhengfang Ding, Riqiang Deng, Dawei Gu. Artificial intelligence accelerate drug discovery [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 278.

Biological activity, molecular docking, and ADME predictions of amphibine analogues of Ziziphus spina-christi towards SARS-CoV-2 Mpro

The main protease of the SARS-CoV-2 virus, SARS-CoV-2 Mpro, can be discovered as a promising target to treat the COVID-19 pandemic. The peptide-based inhibitors may present better options than small molecules for inhibits SARS-CoV-2 Mpro. Ziziphus spina-christi species reported have a peptide-based of alkaloids group, i.e. Amphibine that the analogues can be identified the potential as an inhibitor of SARS-CoV-2 Mpro. The compound structure was drawn and optimized using semi-empirical AM-1 method using Quantum ESPRESSO v.6.6, then the biological activity using PASS Prediction server and molecular docking simulation using MGLTools 1.5.6 with AutoDock 4.2 were performed. Afterward, the ADME profiles were predicted using the SWISS-ADME server. PASS server was predicting Amphibine B-F and H showed potency both as antiviral and as a protease inhibitor. The molecular docking simulation of Amphibine analogues showed lower binding energy than the native ligand. The binding energy of the native ligand was −7.69 Kcal/mol compared to the lowest binding energy of Amphibine analogues was −10.10 Kcal/mol (Amphibine-F). The ADME prediction showed, as an oral drug Amphibine-F has the best bioavailability, Amphibine-B, C, and D have good bioavailability, and Amphibine-E and H have poor bioavailability. Concluded, Amphibine B-F and H of Amphibine analogues showed potency as COVID-19 treatment targeting SARS-CoV-2 Mpro.

Prediction of chemical compounds properties using a deep learning model

The discovery of new medications in a cost-effective manner has become the top priority for many pharmaceutical companies. Despite decades of innovation, many of their processes arguably remain relatively inefficient. One such process is the prediction of biological activity. This paper describes a new deep learning model, capable of conducting a preliminary screening of chemical compounds in-silico. The model has been constructed using a variation autoencoder to generate chemical compound fingerprints, which have been used to create a regression model to predict their LogD property and a classification model to predict binding in selected assays from the ChEMBL dataset. The conducted experiments demonstrate accurate prediction of the properties of chemical compounds only using structural definitions and also provide several opportunities to improve upon this model in the future.

Analysis of the relationship between the predicted biological activity and the chemical structure of S-derivatives of 5-(5-bromofuran-2-yl)-4R-1,2,4-triazole-3-thiols

1,2,4-Triazole derivatives are actively used as components in the development of new drugs, plant protection products, polymeric materials, anti-corrosion agents and etc. Chemical modeling of substituted 1,2,4-triazoles due to the introduction of different pharmacophores into the structure is very popular among scientists in various fields. Today it is known that some S-derivatives of 5-(5-bromofuran-2-yl)-4R-1,2,4-triazole-3-thiols have antimicrobial activity. The aim of the work is to analyze the relationships between the predicted biological activity and the chemical structure of S-derivatives of 5-(5-bromofuran-2-yl) -4R-1,2,4-triazole-3-thiols. Materials and methods. Virtual screening of compounds was performed using the computer program PASS (Prediction of activity spectra for substances). The results of the forecast were issued in the form of a list of names of probable types of activity with estimates of the probabilities of presence (Pa) and absence of each activity (Pi), which had values from 0 to 1. Results. Analyzing the prediction of biological activity on protein targets from the group of enzymes, we can said that derivatives of 5-(5-bromofuran-2-yl)-4R-1,2,4-triazole-3-thiols were active in the group of oxyreductases (Glutathione reductase, mitochondrial; Cyclooxygenase-2; Hypoxia-inducible factor prolyl hydroxylase 2), which catalyzed oxidation reactions, the transfer of electrons from one molecule (reducer, electron donor) to another (oxidant, electron acceptor). These compounds can demonstrate antioxidant, antihypoxic activity. Conclusions. The conducted forecast of biological activity revealed that derivatives of 5-(5-bromofuran-2-yl)-4R-1,2,4-triazole-3-thiols are the most active and there is a probability to show antitumor, antiviral, antibacterial, diuretic, actoprotective, and antioxidant activity.

Targeted Synthesis and Analysis of Biologically Active Azomethine Derivatives of 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide

Introduction. Azomethine derivatives of 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide are acyclic precursors of biologically active compounds derived from 5,6,7,8-tetrahydro-3H-benzoteopheno[2,3-d]pyrimidine-4-one. Examples of these groups of compounds with different pharmacological properties are given in the literature, but their cytostatic effect is mainly described. These data and the preparative availability allow us to judge the prospects for further study and molecular design in a number of azomethine derivatives of 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide. Optimization of methods for the synthesis and analysis of substances of this series and the identification of structure-activity relationship are of considerable interest for medical chemistry and pharmaceutical science. The resulting leading compounds will allow us to further develop laboratory requirements for the synthesis of an active pharmaceutical substance.Aim. To make a predict, optimize the synthesis conditions and develop a method for high performance liquid chromatography (HPLC) analysis of pharmacologically active azomethine derivatives of 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide.Materials and methods. The prediction of biological activity was carried out through the web resource PASS Online. The synthesis of the target azomethines was carried out by the interaction of aromatic aldehydes with 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide in an ethanol. The reaction was monitored by thin-layer chromatography (TLC). The determination of related impurities was done by HPLC. The analysis was carried out under the conditions of isocratic elution with a mobile phase of acetonitrile – water (70:30).Results and discussion. The results of the prediction of the biological activity of the constructed structures suggest the manifestation of cytostatic, antitubercular and anti-inflammatory activity characteristic of all target azomethines. The analysis of the reactivity revealed the influence of substituents of aldehydes contained in the aromatic core on the completeness of the condensation reaction. The spectral characteristics clearly confirmed the structure of the products, and the HPLC results showed the purity of the obtained substances, which is more than 95 %.Conclusion. As a result of the conducted studies, the structure of promising azomethine derivatives of 2-amino-4,5,6,7-tetrahydro-1-benzothiophene-3-carboxamide was justified and the method of their synthesis and analysis by HPLC was optimized. In the future, the results of the research will allow us to identify the leading compounds with the specified pharmacological properties.

A Machine Learning Bioinformatics Method to Predict Biological Activity from Biosynthetic Gene Clusters.

Research in natural products, the genetically encoded small molecules produced by organisms in an idiosyncratic fashion, deals with molecular structure, biosynthesis, and biological activity. Bioinformatics analyses of microbial genomes can successfully reveal the genetic instructions, biosynthetic gene clusters, that produce many natural products. Genes to molecule predictions made on biosynthetic gene clusters have revealed many important new structures. There is no comparable method for genes to biological activity predictions. To address this missing pathway, we developed a machine learning bioinformatics method for predicting a natural product’s antibiotic activity directly from the sequence of its biosynthetic gene cluster. We trained commonly used machine learning classifiers to predict antibacterial or antifungal activity based on features of known natural product biosynthetic gene clusters. We have identified classifiers that can attain accuracies as high as 80% and that have enabled the identification of biosynthetic enzymes and their corresponding molecular features that are associated with antibiotic activity.

Virtual screening of natural compounds for potential inhibitors of Sterol C-24 methyltransferase of Leishmania donovani to overcome leishmaniasis.

Leishmaniasis is a neglected tropical disease caused by trypanosomatid parasite belonging to the genera Leishmania. Leishmaniasis is transmitted from one human to other through the bite of sandflies. It is endemic in around 98 countries including tropical and subtropical regions of Asia, Africa, Southern America, and the Mediterranean region. Sterol C-24 methyltransferase (LdSMT) of Leishmania donovani (L. donovani) mediates the transfer of CH3-group from S-adenosyl methionine to C-24 position of sterol side chain which makes the ergosterol different from cholesterol. Absence of ortholog in human made it potential druggable target. Here, we performed virtual screening of library of natural compounds against LdSMT to identify the potential inhibitor for it and to fight leishmaniasis. Gigantol, flavan-3-ol, and parthenolide showed the best binding affinity towards LdSMT. Further, based on absorption, distribution, metabolism, and excretion properties and biological activity prediction, gigantol showed the best lead-likeness and drug-likeness properties. Therefore, we further elucidated its antileishmanial properties. We found that gigantol inhibited the growth and proliferation of promastigotes as well as intra-macrophagic amastigotes. Gigantol exerted its antileishmanial action through the induction of reactive oxygen species in dose-dependent manner. Our study, suggested the possible use of gigantol as antileishmanial drug after further validations to overcome leishmaniasis.

Synthesis, crystal structure, Hirschfeld surface analysis and biological activity prediction of N-(dimethoxyphosphoryl)-1-methylpyridinium-4-carboximidate

N-(dimethoxyphosphoryl)-1-methylpyridinium-4-carboximidate, a new carbacylamidophosphate-type compound, was synthesized and characterized by means of IR, 1H and 31P NMR and UV-Vis spectroscopies and X-ray analysis. The molecule of the synthesized compound has triclinic (P-1) symmetry, displays monomeric motif in crystal and crystalizes as solvate containing methanol molecule, which is connected to carbacylamidophosphate molecule through O(2)H(5A)–O(5) hydrogen bond. Through – stacking interactions, the molecules of the synthesized compound are linked in the chain along the a crystallographic axis. Several other intermolecular bonds connect these chains along b and c crystallographic axes. The intermolecular interactions with HH and OH contacts prevail in the crystalline structure of N-(dimethoxyphosphoryl)-1-methylpyridinium-4-carboximidate, the contribution of planar stacking CC contacts being equal to 4.1%. The synthesized compound was found to be well soluble in water. By using computer program PASS, we established that the synthesized substance is likely can exhibit 18 types of biological activity in experiment.

Quantum Chemical Studies on Structural, Spectroscopic, Thermochemistry, Photo-physical and Bioactivity Properties of m-Cresol Purple Dye

The m-Cresol purple molecule is analyzed using spectroscopy and quantum computational chemistry methods using the software program Gaussian 09. B3LYP 6-311G (d, p) level has been used to create stable conformation of molecular structure, vibrational frequencies, Mulliken atomic charges, and electronic absorption spectra. The active regions of the infrared intensities, polarizabilities, and first-order polarizabilities were determined. The visible ultrasonic absorption with DOS spectrum demonstrated the highest correlation both before and after UV exposure. Furthermore, ’Frontier’s molecular orbital analysis was determined, explaining the difference between HOMOs and LUMOs energies. Swiss ADME is used to measure physicochemical descriptors and the prediction of molecular dynamics, ADME (absorption, distribution, metabolism, excretion) coefficients, pharmaco-kinetics pH, log P, biological activity, and drug-like nature. Furthermore, the predictive model of BOILED-Egg, QSAR analysis, molecular lipophilicity, distribution of microorganisms, target binding percentages, and topology measurements are analyzed to help drug discovery.

Chemical studies of some fused heterocyclic compounds containing oxygen with anticipated biological interests: Thesis Abstract

Thesis AbstractBackground: Heterocyclic are known to have multiple applications in particular biomedical settings. Aim: The present study reports a convenient protocol for the design and synthesis of new heterocyclic compounds incorporating 1- hydroxy-naphthyl and new angular chromones bearing allylnaphthyl fragments. Materials and Methods: The versatile precursor 2-acetyl-4- allyl-1-hydroxy naphthalene was subjected to Claisen-Schmidt condensation but the precursor 6-allyl-3-formyl-4H-benzo[h]chromone was prepared via Vilsmeier–Haack reaction of 4-allyl-1- hydroxy-2-acetonaphthone. Results: Some new compounds are considered as promising leads with dual biological activity as antibacterial and antioxidant. The isoalloxazine ring of the cofactor FAD is essential to the recognition of the compound that has proper antioxidant activity. Benzocoumarin ring is essential for the recognition of the amino acid residues. Allyl chain improves the p stacking interaction while the n-propyl chain attenuates the overall interaction. Naphthyl ring needs to be substituted with a short bridge to avid the ring extended away from the pocket. The antimicrobial and antiquorum-sensing activities of the newly synthesized compounds showed that many compounds have good and promising activities in comparison with the reference drugs. Conclusion: Based on the computational prediction of pharmacokinetics, drug-likeness properties, biological activity, and molecular docking suggested that formyl chromone and tetracarbonyl may be potent antimicrobial drugs

Chemical Studies on Some Chromone Derivatives with Anticipated Pharmaceutical Applications: Thesis Abstract

The present study reports a convenient protocol for the design and synthesis of new heterocyclic compounds incorporating 1- hydroxy-naphthyl and new angular chromones bearing allylnaphthyl fragments. The versatile precursor 2-acetyl-4- allyl-1-hydroxy naphthalene was subjected to Claisen-Schmidt condensation but The precursor 6-allyl-3-formyl-4H-benzo[h]chromone was prepared via Vilsmeier–Haack reaction of 4-allyl-1- hydroxy-2-acetonaphthone.Some new compounds are considered as promising leads with dual biological activity as antibacterial and antioxidant. The isoalloxazine ring of the cofactor FAD is essential to the recognition of the compound that has proper antioxidant activity. Benzocoumarin ring is essential for the recognition of the amino acid residues. Allyl chain improves the p stacking interaction while the n-propyl chain attenuates the overall interaction. Naphthyl ring needs to be substituted with a short bridge to avid the ring extended away from the pocket.The antimicrobial and antiquorum-sensing activities of the newly synthesized compounds showed that many compounds have good and promising activities in comparison with the reference drugs. Computational prediction of pharmacokinetics, drug-likeness properties, biological activity, and molecular docking suggested that formyl chromone and tetracarbonyl may be potent antimicrobial drugs.

Hepatic Alveolar Echinococcosis: Predictive Biological Activity Based on Radiomics of MRI.

Background To evaluate the role of radiomics based on magnetic resonance imaging (MRI) in the biological activity of hepatic alveolar echinococcosis (HAE). Methods In this study, 90 active and 46 inactive cases of HAE patients were analyzed retrospectively. All the subjects underwent MRI and positron emission tomography computed tomography (PET-CT) before surgery. A total of 1409 three-dimensional radiomics features were extracted from the T2-weighted MR images (T2WI). The inactive group in the training cohort was balanced via the synthetic minority oversampling technique (SMOTE) method. The least absolute shrinkage and selection operator (LASSO) regression method was used for feature selection. The machine learning (ML) classifiers were logistic regression (LR), multilayer perceptron (MLP), and support vector machine (SVM). We used a fivefold cross-validation strategy in the training cohorts. The classification performance of the radiomics signature was evaluated using receiver operating characteristic curve (ROC) analysis in the training and test cohorts. Results The radiomics features were significantly associated with the biological activity, and 10 features were selected to construct the radiomics model. The best performance of the radiomics model for the biological activity prediction was obtained by MLP (AUC = 0.830 ± 0.053; accuracy = 0.817; sensitivity = 0.822; specificity = 0.811). Conclusions We developed and validated a radiomics model as an adjunct tool to predict the HAE biological activity by combining T2WI images, which achieved results nearly equal to the PET-CT findings.

Evaluation of Pipemidic Acid Derivatives for Potential Antimicrobial Activity Application: In silico Studies on Bioactivity

Background: Pipemidic acid is a broad-spectrum quinolone antibacterial agent for the treatment of chronic urinary tract infections against both gram-positive and gram-negative bacteria. Both quinolone and fluoroquinolone antibiotics have been useful in combating bacterial infections. However, patients suffer severe side effects when they stop taking the medication. The piperazinyl region of pipemidic acid is highly responsible for the side effects. Objective: The objective of this study is to design new compounds in which the piperazinyl region is masked by way of conjugation to benzoic acid derivatives. Methods: In silico studies were conducted using AutoDockTools software for ligand-protein docking. The docking scores were compared to the parent pipemidic acid docked to Bacterial DNA (deoxyribonucleic acid) gyrase and GABA (gamma-Aminobutyric acid) receptor from the PDB (Protein Data Bank) database. Sites of metabolism, biological activity, quantum chemical descriptors, and ADME (absorption, distribution, metabolism, and excretion) property predictions for each designed ligand were also evaluated. Results: The docking studies and biological activity predictions showed good anti-infective properties (ligand PAR03) whilst also suggesting a reduction in GABA receptor agonist activity. The performance of PAR03 correlates with its electronic properties showing electrophilic character (can generate Reactive Electrophilic Species (RES)). Conclusion: The results from this study indicate that modification of the piperazinyl region of pipemidic acid can be an effective way to improve the drug potency whilst also ensuring reduction of the associated side effects.

Repurposing of approved drug molecules for viral infectious diseases: a molecular modelling approach

The identification of new viral drugs has become a task of paramount significance due to the frequent occurrence of viral infections and especially during the current pandemic. Despite the recent advancements, the development of antiviral drugs has not made parallel progress. Reduction of time frame and cost of the drug development process is the major advantage of drug repurposing. Therefore, in this study, a drug repurposing strategy using molecular modelling techniques, i.e. biological activity prediction, virtual screening, and molecular dynamics simulation was employed to find promising repurposing candidates for viral infectious diseases. The biological activities of non-redundant (4171) drug molecules were predicted using PASS analysis, and 1401 drug molecules were selected which showed antiviral activities in the analysis. These drug molecules were subjected to virtual screening against the selected non-structural viral proteins. A series of filters, i.e. top 10 drug molecules based on binding affinity, mean value of binding affinity, visual inspection of protein-drug complexes, and number of H-bond between protein and drug molecules were used to narrow down the drug molecules. Molecular dynamics simulation analysis was carried out to validate the intrinsic atomic interactions and binding conformations of protein-drug complexes. The binding free energies of drug molecules were assessed by employing MMPBSA analysis. Finally, nine drug molecules were prioritized, as promising repurposing candidates with the potential to inhibit the selected non-structural viral proteins. Communicated by Ramaswamy H. Sarma

Use of Artificial Intelligence and Machine Learning for Discovery of Drugs for Neglected Tropical Diseases.

Neglected tropical diseases continue to create high levels of morbidity and mortality in a sizeable fraction of the world’s population, despite ongoing research into new treatments. Some of the most important technological developments that have accelerated drug discovery for diseases of affluent countries have not flowed down to neglected tropical disease drug discovery. Pharmaceutical development business models, cost of developing new drug treatments and subsequent costs to patients, and accessibility of technologies to scientists in most of the affected countries are some of the reasons for this low uptake and slow development relative to that for common diseases in developed countries. Computational methods are starting to make significant inroads into discovery of drugs for neglected tropical diseases due to the increasing availability of large databases that can be used to train ML models, increasing accuracy of these methods, lower entry barrier for researchers, and widespread availability of public domain machine learning codes. Here, the application of artificial intelligence, largely the subset called machine learning, to modelling and prediction of biological activities and discovery of new drugs for neglected tropical diseases is summarized. The pathways for the development of machine learning methods in the short to medium term and the use of other artificial intelligence methods for drug discovery is discussed. The current roadblocks to, and likely impacts of, synergistic new technological developments on the use of ML methods for neglected tropical disease drug discovery in the future are also discussed.

A Theoretical Study of 2-hydroxyethyl Substituted NHC Precursors Containing ortho–, meta– and para– methylbenzyl: Global Reactivity Descriptors and Prediction of Biological Activities

In this work, three 2-hydroxyethyl substituted N-heterocyclic carbene (NHC) precursors containing ortho–, meta– and para– methylbenzyl fragments are characterized theoretically. Theoretical calculations are performed to gain insight into these three molecules’ electronic properties (HOMO-LUMO energy, MEP and global reactivity descriptors) and biological behaviors. Also, atomic charges are calculated and molecular orbital analysis is performed. In order to investigate the stability of the molecules resulting from hyperconjugative interactions and charge delocalization, natural bond orbital (NBO) analysis is used. A predictive study for the biological activities is carried out using PASS (prediction of activity spectra for biologically active structures) online software. Biological activity predictions showed the substance P antagonist, analgesic and antiinflammatory activities of the compounds.

Evaluating Deep Learning models for predicting ALK-5 inhibition.

Computational methods have been widely used in drug design. The recent developments in machine learning techniques and the ever-growing chemical and biological databases are fertile ground for discoveries in this area. In this study, we evaluated the performance of Deep Learning models in comparison to Random Forest, and Support Vector Regression for predicting the biological activity (pIC50) of ALK-5 inhibitors as candidates to treat cancer. The generalization power of the models was assessed by internal and external validation procedures. A deep neural network model obtained the best performance in this comparative study, achieving a coefficient of determination of 0.658 on the external validation set with mean square error and mean absolute error of 0.373 and 0.450, respectively. Additionally, the relevance of the chemical descriptors for the prediction of biological activity was estimated using Permutation Importance. We can conclude that the forecast model obtained by the deep neural network is suitable for the problem and can be employed to predict the biological activity of new ALK-5 inhibitors.