Search

Cholesterol oxidases (ChOxes) are enzymes that catalyze the oxidation of cholesterol to cholest-4-en-3-one. These enzymes find wide applications across various diagnostic and industrial settings. In addition, as a pathogenic factor of several bacteria, they have significant clinical implications. The current classification system for ChOxes is based on the type of bond connecting FAD to the apoenzyme, which does not adequately illustrate the enzymatic and structural characteristics of these proteins. In this study, we have adopted an integrative approach, combining evolutionary analysis, classic enzymatic techniques and computational approaches, to elucidate the distinct features of four various ChOxes from Rhodococcus sp. (RCO), Cromobacterium sp. (CCO), Pseudomonas aeruginosa (PCO) and Burkhoderia cepacia (BCO). Comparative and evolutionary analysis of substrate-binding domain (SBD) and FAD-binding domain (FBD) helped to reveal the origin of ChOxes. We discovered that all forms of ChOxes had a common ancestor and that the structural differences evolved later during divergence. Further examination of amino acid variations revealed SBD as a more variable compared to FBD independently of FAD coupling mechanism. Revealed differences in amino acid positions turned out to be critical in determining common for ChOxes properties and those that account for the individual differences in substrate specificity. A novel look with the help of chemical descriptors on found distinct features were sufficient to attempt an alternative classification system aimed at application approach. While univocal characteristics necessary to establish such a system remain elusive, we were able to demonstrate the substrate and protein features that explain the differences in substrate profile. [Display omitted] • Current classification of cholesterol oxidases is useless for enzymatic description. • All forms of ChOxes had a common ancestor with further divergence. • Chemical descriptors turned out to be a suitable for alternative classification. • Folding-based approach can better guide the selection of enzymes for practice. [ABSTRACT FROM AUTHOR]

This study enhances the molecular analysis of bitumen by transitioning from traditional chemical descriptors, such as SARA (Saturates, Aromatics, Resins, and Asphaltenes) fractions and elemental compositions, to specific force field atom types in Molecular Dynamics (MD) models. This shift improves the precision in predicting material properties critical for bituminous material characterization. Machine Learning Models (MLMs) were developed to use these atom types as input features, inherently reflecting fundamental chemical characteristics. Trained on data from over 1,770 LAMMPS simulations of diverse bitumen types and conditions, these MLMs enable the prediction of properties like density, heat capacity, solubility parameters, and thermal expansion coefficients without the need for additional MD simulations. The models utilize 30 chemical descriptors corresponding to specific atom types in the PCFF force field, which collectively account for over 95% of the influence on these properties. By accurately predicting fundamental, thermodynamic, and kinetic properties, the use of MLMs and force field atom types allows researchers to efficiently tweak the chemical nature of organic molecules and mixtures to achieve desired properties. With near-instantaneous prediction times, these MLMs offer valuable insights for advancing bitumen research in the construction and petroleum industries, reducing the need for more intensive simulation techniques.

The physicochemical properties of liquid alternative fuels are important but difficult to measure/predict, especially when complex surrogate fuels are concerned. In the present work, machine learning is used to develop quantitative structure–property relationship models. The fuel chemical structure is represented by molecular descriptors, allowing the linking of important features of the fuel composition and key properties of fuel utilization. Feature selection is employed to select the most relevant features that describe the chemical structure of the fuel and several machine learning algorithms are tested to construct interpretable models. The effectiveness of the methodology is demonstrated through the development of accurate and interpretable predictive models for cetane numbers, with a focus on understanding the link between molecular structure and fuel properties. In this context, matrix-based descriptors and descriptors related to the number of atoms in the molecule are directly linked with the cetane number of hydrocarbons. Furthermore, the results showed that molecular connectivity indices play a role in the cetane number for aromatic molecules. Also, the methodology is extended to predict the cetane number of ester and ether molecules, leveraging the design of alternative fuels towards fully sustainable fuel utilization.

Density functional theory (DFT) calculations and molecular docking have been carried out on natural products containing eugenol, gingerol, ascorbic acid, oleurpoein, piperine, hesperidin, quercetin, Luteolin, and curcumin in order to predict their biological activities and to analyze their pancreatic lipase inhibition. The biological activity predictions are based on the global and local chemical descriptors, namely, HOMO–LUMO gaps, chemical hardness, chemical potential, electrophilicity, dipole moment, and Fukui functions. Our findings show that the studied compounds can be divided into two groups based on the chemical descriptors; the first group is composed of eugenol, gingerol, ascorbic acid, and oleuropein and the second one is composed of piperine, hesperidin, quercetin, Luteolin, and curcumin depending on the HOMO–LUMO gaps and electrophilicity values predicting best reactivity for the second group than the first one. The frontier orbitals offer a deeper insight concerning the electron donor and electron acceptor capabilities, whereas the local descriptors resulting from Fukui functions put emphasis on the active sites of different candidate ligands. The molecular docking was performed in order to compare and identify the inhibition activity of the natural candidate ligands against pancreatic lipase which were compared to that of synthesized ones. The molecular docking results revealed that the Luteolin compound has the best binding affinity of −8.56 kcal/mol due to their unique molecular structure and the position of -OH aromatic substituents. [ABSTRACT FROM AUTHOR]

The application of artificial intelligence to chemistry usually focuses on the identification of good correlations between descriptors and a given property of interest. The descriptors often come from arbitrary sets, with the implicit assumption that the evaluation of a sufficiently wide range of descriptors will lead to a satisfactory choice. Recent work in our group has focused on applying statistical analysis to large amounts of DFT results with the goal of finding optimal descriptor sets for a given property, which we label as hidden descriptors. This article briefly discusses this treatment and the chemical knowledge that has been gained through its application in two different domains: metal-ligand bond strength in transition metal complexes, and energy barriers in bimolecular nucleophilic substitution reactions.

Machine learning has proven to be a powerful tool for accelerating biofuel development. Although numerous models are available to predict a range of properties using chemical descriptors, there is a trade-off between interpretability and performance. Neural networks provide predictive models with high accuracy at the expense of some interpretability, while simpler models such as linear regression often lack in accuracy. In addition to model architecture, feature selection is also critical for developing interpretable and accurate predictive models. We present a method for systematically selecting molecular descriptor features and developing interpretable machine learning models without sacrificing accuracy. Our method simplifies the process of selecting features by reducing feature multicollinearity and enables discoveries of new relationships between global properties and molecular descriptors. To demonstrate our approach, we developed models for predicting melting point, boiling point, flash point, yield sooting index, and net heat of combustion with the help of the Tree-based Pipeline Optimization Tool (TPOT). For training, we used publicly available experimental data for up to 8351 molecules. Our models accurately predict various molecular properties for organic molecules (mean absolute percent error (MAPE) ranges from 3.3% to 10.5%) and provide a set of features that are well-correlated to the property. This method enables researchers to explore sets of features that significantly contribute to the prediction of the property, offering new scientific insights. To help accelerate early stage biofuel research and development, we also integrated the data and models into a open-source, interactive web tool.

Coriandrum sativum L. and Foeniculum vulgare Mill. are two aromatic and medicinal Apiaceae species commonly grown in Portugal, whose essential oils (EOs) are used in the food, pharmaceutical, and cosmetics industries. The present study evaluated EOs isolated from the fruits and vegetative aerial parts (VAPs) of 11 samples of Coriandrum sativum L. and from the fruits of 19 samples of Foeniculum vulgare Mill. The plant material was grown in experimental fields, after collection from several regions of mainland Portugal. The EOs were isolated by hydrodistillation and analyzed by gas chromatography and gas chromatography–mass spectrometry. The coriander EOs analysis evidenced two main clusters, with the first containing the fruits' EOs dominated by linalool (60–73%), γ-terpinene, and α-pinene and the second with the VAPs' EOs, which showed 2-trans-decenal (37–63%) and n-decanal (13–30%) as the main compounds. The fennel EOs analysis revealed two well correlated clusters, the first dominated by estragole (34–76%) and fenchone (16–30%) and the other dominated by trans-anethole (37–56%) and fenchone (14–34%). The present data suggest coriander EOs' chemical descriptors as linalool for the fruits' EOs and 2-trans-decenal with n-decanal for the VAPs' EOs. For the fennel fruit EOs, the putative descriptors were trans-anethole and estragole, with variable fenchone content. The gathered data reinforce the relevance of clarifying variability of these species' EOs, particularly when considering aromatic and medicinal plants with such a wide range of applications. [ABSTRACT FROM AUTHOR]

Origanum vulgare L. is a well-known aromatic and medicinal plant, whose essential oil (EO) has recognised flavouring and medicinal properties. In this study, Origanum vulgare subsp. virens (Hoffmanns. & Link) Ietsw. EOs, isolated from accessions grown in experimental fields, were evaluated. The plant material was grown from rooted cuttings or nutlets (fruits), originally collected in 20 regions in mainland Portugal and harvesting for EO isolation was performed in two years. EOs were isolated by hydrodistillation and analysed by gas chromatography and gas chromatography–mass spectrometry, for EO quantification and identification, respectively. EO yields ranged from <0.05–3.3% for rooted cuttings, with oregano samples obtained in Portalegre and Alandroal, respectively. Ninety-one compounds were identified, mainly grouped in oxygen-containing monoterpenes and monoterpene hydrocarbons. EO agglomerative cluster analysis evidenced two main clusters, with the first subdivided into four subclusters. From the obtained data, the putative O. vulgare subsp. virens chemotypes are carvacrol, thymol and linalool, with γ-terpinene, p-cymene, cis- and trans-β-ocimene also contributing as these EOs chemical descriptors. The comparison between the present data and a survey of the existing literature on Portuguese O. vulgare reinforces the major variability of this species' EOs and emphasises the importance of avoiding wild collections to obtain a defined chemical type of crop production of market relevance. [ABSTRACT FROM AUTHOR]

A heteroaryl sulfonyl(VI) fluoride, 4‐chloro‐7‐fluorosulfonyl‐2,1,3‐benzoxadiazole, was synthesized from its chloride counterpart (4‐chloro‐7‐chlorosulfonyl‐2,1,3‐benzoxadiazole) and the X‐ray structure analysis of these compounds and the interactions in the solid‐state were thoroughly examined. Hirshfeld surface analysis is used to provide a thorough and complete picture of the changes arising from the different halides in the functional groups. Surface analysis reveals that the fluoride does not participate in any hydrogen interactions as opposed to the chloride. However, the fluorine atom is observed to form close interactions with several π bonds. For both moieties, however, the sulfonyl oxygens show comparable interactions with respect to both magnitude and interatomic distances. The Hirshfeld surface analysis is coupled with computational studies to help elucidate the observed interactions that are found from the distinct nitrogen, chlorine, and oxygen atoms present in the molecules, providing new physical insights to the correlation between their structures and properties [ABSTRACT FROM AUTHOR]

[Display omitted] • Force field atom types accurately characterize chemical impact on bitumen properties, surpassing conventional methods. • Machine Learning Models (MLMs) predict bitumen properties, reducing the need for extensive molecular simulations. • Examined 30 chemical descriptors: 10 key features influence over 95% of bitumen material properties. • Near instantaneous property estimation allows for rapid evaluation of bitumen quality and performance. This study enhances the molecular analysis of bitumen by transitioning from traditional chemical descriptors, such as SARA (Saturates, Aromatics, Resins, and Asphaltenes) fractions and elemental compositions, to specific force field atom types in Molecular Dynamics (MD) models. This shift improves the precision in predicting material properties critical for bituminous material characterization. Machine Learning Models (MLMs) were developed to use these atom types as input features, inherently reflecting fundamental chemical characteristics. Trained on data from over 1,770 LAMMPS simulations of diverse bitumen types and conditions, these MLMs enable the prediction of properties like density, heat capacity, solubility parameters, and thermal expansion coefficients without the need for additional MD simulations. The models utilize 30 chemical descriptors corresponding to specific atom types in the PCFF force field, which collectively account for over 95% of the influence on these properties. By accurately predicting fundamental, thermodynamic, and kinetic properties, the use of MLMs and force field atom types allows researchers to efficiently tweak the chemical nature of organic molecules and mixtures to achieve desired properties. With near-instantaneous prediction times, these MLMs offer valuable insights for advancing bitumen research in the construction and petroleum industries, reducing the need for more intensive simulation techniques. [ABSTRACT FROM AUTHOR]

Pyrimidine compounds have proven to be effective and efficient additives capable of protecting mild steel in acidic media. This class of organic compounds often functions as adsorption-type inhibitors of corrosion by forming a protective layer on the metallic substrate. The present study reports a computational study of forty pyrimidine compounds that have been investigated as sustainable inhibitors of mild steel corrosion in molar HCl solution. Quantitative structure property relationship was conducted using linear (multiple linear regression) and nonlinear (artificial neural network) models. Standardization method was employed in variable selection yielding five top chemical descriptors utilized for model development along with the inhibitor concentration. Multiple linear regression model yielded a fair predictive model. Artificial neural network model developed using k-fold cross-validation method provided a comprehensive insight into the corrosion protection mechanism of studied pyrimidine-based corrosion inhibitors. Using a multilayer perceptron with Levenberg–Marquardt algorithm, the study obtained the optimal model having a MSE of 8.479, RMSE of 2.912, MAD of 1.791, and MAPE of 2.648. The optimal neural network model was further utilized to forecast the protection capacities of nine non-synthesized pyrimidine derivatives. The predicted inhibition efficiencies ranged from 89 to 98%, revealing the significance of the considered chemical descriptors, the predictive capacity of the developed model, and the potency of the theoretical inhibitors. [ABSTRACT FROM AUTHOR]

Antibiotic drug resistance has emerged as a major public health threat globally. One of the leading causes of drug resistance is the colonization of microorganisms in biofilm mode. Hence, there is an urgent need to design novel and highly effective biofilm inhibitors that can work either synergistically with antibiotics or individually. Therefore, we have developed a recursive regression-based platform "Biofilm-i" employing a quantitative structure–activity relationship approach for making generalized predictions, along with group and species-specific predictions of biofilm inhibition efficiency of chemical(s). The platform encompasses eight predictors, three analysis tools, and data visualization modules. The experimentally validated biofilm inhibitors for model development were retrieved from the "aBiofilm" resource and processed using a 10-fold cross-validation approach using the support vector machine and andom forest machine learning techniques. The data was further sub-divided into training/testing and independent validation sets. From training/testing data sets the Pearson's correlation coefficient of overall chemicals, Gram-positive bacteria, Gram-negative bacteria, fungus, Pseudomonas aeruginosa, Staphylococcus aureus, Candida albicans, and Escherichia coli was 0.60, 0.77, 0.62, 0.77, 0.73, 0.83, 0.70, and 0.71 respectively via Support Vector Machine. Further, all the QSAR models performed equally well on independent validation data sets. Additionally, we also checked the performance of the random forest machine learning technique for the above datasets. The integrated analysis tools can convert the chemical structure into different formats, search for a similar chemical in the aBiofilm database and design the analogs. Moreover, the data visualization modules check the distribution of experimentally validated biofilm inhibitors according to their common scaffolds. The Biofilm-i platform would be of immense help to researchers engaged in designing highly efficacious biofilm inhibitors for tackling the menace of antibiotic drug resistance. [ABSTRACT FROM AUTHOR]

Coriandrum sativum L. and Foeniculum vulgare Mill. are two aromatic and medicinal Apiaceae species commonly grown in Portugal, whose essential oils (EOs) are used in the food, pharmaceutical, and cosmetics industries. The present study evaluated EOs isolated from the fruits and vegetative aerial parts (VAPs) of 11 samples of Coriandrum sativum L. and from the fruits of 19 samples of Foeniculum vulgare Mill. The plant material was grown in experimental fields, after collection from several regions of mainland Portugal. The EOs were isolated by hydrodistillation and analyzed by gas chromatography and gas chromatography–mass spectrometry. The coriander EOs analysis evidenced two main clusters, with the first containing the fruits’ EOs dominated by linalool (60–73%), γ-terpinene, and α-pinene and the second with the VAPs’ EOs, which showed 2-trans-decenal (37–63%) and n-decanal (13–30%) as the main compounds. The fennel EOs analysis revealed two well correlated clusters, the first dominated by estragole (34–76%) and fenchone (16–30%) and the other dominated by trans-anethole (37–56%) and fenchone (14–34%). The present data suggest coriander EOs’ chemical descriptors as linalool for the fruits’ EOs and 2-trans-decenal with n-decanal for the VAPs’ EOs. For the fennel fruit EOs, the putative descriptors were trans-anethole and estragole, with variable fenchone content. The gathered data reinforce the relevance of clarifying variability of these species’ EOs, particularly when considering aromatic and medicinal plants with such a wide range of applications.

The covalent and noncovalent backbone binding interactions in RNA‐peptide complexes were studied by DFT methods. Four RNA structures R1(GGCUAGCC), R2(AAUCGAUU), R3(GGGAUCCC), and R4(AAAGCUUU) has been selected for eight protonated peptides (DR, ER, GR, KR, NGR, RR, tmeGnd (tme), and VR) interactions based on an experimental study (Anal Chem. 2019; 91:1659‐1664). Chemical reactivity theory is used to study the reactivity of eight peptides with global descriptors. Lower hardness values reflected low stability and high reactivity for the protonated peptides. DR, ER, GR, KR, NGR, RR, and VR show lower value of ω, μ while tme has high value of ω, μ. Larger highest occupied molecular orbital (HOMO)–lowest unoccupied molecular orbital (LUMO) gap for ER, GR, and KR showed greater structural stability for peptides. AutoDock and PatchDock results indicated that R1, R2, and R4 retain hairpin structures while interacting with peptide complexes. The calculated binding energies of (R1–R4)‐peptide complexes from AutoDock tools are (1.49–11.12) kcal/mol. Results showed that the noncovalent interactions are stronger than the covalent interactions for R1‐peptide complexes. The reason might be the transfer of proton from protonated ligand to deprotonated RNA, which has initiated the loss of the ligand. Also it has been observed that proton transfer has become energetically unfavorable in presence of additional hydrogen bonds which is predicted in the experimental results. [ABSTRACT FROM AUTHOR]

There has recently been a rapid progress in computational methods for determining protein targets of small molecule drugs, which will be termed as compound protein interaction (CPI). In this review, we comprehensively review topics related to computational prediction of CPI. Data for CPI has been accumulated and curated significantly both in quantity and quality. Computational methods have become powerful ever to analyze such complex the data. Thus, recent successes in the improved quality of CPI prediction are due to use of both sophisticated computational techniques and higher quality information in the databases. The goal of this article is to provide reviews of topics related to CPI, such as data, format, representation, to computational models, so that researchers can take full advantages of these resources to develop novel prediction methods. Chemical compounds and protein data from various resources were discussed in terms of data formats and encoding schemes. For the CPI methods, we grouped prediction methods into five categories from traditional machine learning techniques to state-of-the-art deep learning techniques. In closing, we discussed emerging machine learning topics to help both experimental and computational scientists leverage the current knowledge and strategies to develop more powerful and accurate CPI prediction methods.

Origanum vulgare L. is a well-known aromatic and medicinal plant, whose essential oil (EO) has recognised flavouring and medicinal properties. In this study, Origanum vulgare subsp. virens (Hoffmanns. & Link) Ietsw. EOs, isolated from accessions grown in experimental fields, were evaluated. The plant material was grown from rooted cuttings or nutlets (fruits), originally collected in 20 regions in mainland Portugal and harvesting for EO isolation was performed in two years. EOs were isolated by hydrodistillation and analysed by gas chromatography and gas chromatography–mass spectrometry, for EO quantification and identification, respectively. EO yields ranged from O. vulgare subsp. virens chemotypes are carvacrol, thymol and linalool, with γ-terpinene, p-cymene, cis- and trans-β-ocimene also contributing as these EOs chemical descriptors. The comparison between the present data and a survey of the existing literature on Portuguese O. vulgare reinforces the major variability of this species’ EOs and emphasises the importance of avoiding wild collections to obtain a defined chemical type of crop production of market relevance.

The retention time (RT) of contaminants of emerging concern (CECs) in liquid chromatography-high-resolution mass spectrometry (LC-HRMS) is crucial for database matching in non-targeted screening (NTS) analysis. In this study, we developed a machine learning (ML) model to predict RTs of CECs in NTS analysis. Using 1051 CEC standards, we evaluated Random Forest (RF), XGBoost, Support Vector Regression (SVR), and Artificial Neural Network (ANN) with molecular fingerprints and chemical descriptors to establish an optimal model. The SVR model utilizing chemical descriptors resulted in good predictive capacity with R 2 ext = 0.850 and r 2 = 0.925. The model was further validated through laboratory NTS compound characterization. When applied to examine CEC occurrence in a large wastewater treatment plant, we identified 40 level S1 CECs (confirmed structure by reference standard) and 234 level S2 compounds (probable structure by library spectrum match). The model predicted RTs for level S2 compounds, leading to the classification of 153 level S2 compounds with high confidence (ΔRT <2 min). The model served as a robust filtering mechanism within the analytical framework. This study emphasizes the importance of predicted RTs in NTS analysis and highlights the potential of prediction models. Our research introduces a workflow that enhances NTS analysis by utilizing RT prediction models to determine compound confidence levels. [Display omitted] • Applied machine learning to create a model predicting CEC retention times. • Assessed four algorithms and different feature sets using 1051 CEC standards. • Lab validation of model showed high confidence in compounds identification. • Introduced workflow using RT prediction model for enhancing compound confidence. • Model applied to wastewater plant identified 153 level S2 CECs in high confidence. [ABSTRACT FROM AUTHOR]

Antibiotic drug resistance has emerged as a major public health threat globally. One of the leading causes of drug resistance is the colonization of microorganisms in biofilm mode. Hence, there is an urgent need to design novel and highly effective biofilm inhibitors that can work either synergistically with antibiotics or individually. Therefore, we have developed a recursive regression-based platform “Biofilm-i” employing a quantitative structure–activity relationship approach for making generalized predictions, along with group and species-specific predictions of biofilm inhibition efficiency of chemical(s). The platform encompasses eight predictors, three analysis tools, and data visualization modules. The experimentally validated biofilm inhibitors for model development were retrieved from the “aBiofilm” resource and processed using a 10-fold cross-validation approach using the support vector machine and andom forest machine learning techniques. The data was further sub-divided into training/testing and independent validation sets. From training/testing data sets the Pearson’s correlation coefficient of overall chemicals, Gram-positive bacteria, Gram-negative bacteria, fungus, Pseudomonas aeruginosa, Staphylococcus aureus, Candida albicans, and Escherichia coli was 0.60, 0.77, 0.62, 0.77, 0.73, 0.83, 0.70, and 0.71 respectively via Support Vector Machine. Further, all the QSAR models performed equally well on independent validation data sets. Additionally, we also checked the performance of the random forest machine learning technique for the above datasets. The integrated analysis tools can convert the chemical structure into different formats, search for a similar chemical in the aBiofilm database and design the analogs. Moreover, the data visualization modules check the distribution of experimentally validated biofilm inhibitors according to their common scaffolds. The Biofilm-i platform would be of immense help to researchers engaged in designing highly efficacious biofilm inhibitors for tackling the menace of antibiotic drug resistance.

Density functional theory (DFT) was performed in order to predict the structural, chemical descriptors and optoelectronic properties of the drugs Hydroxychloroquine and Azithromycin using the wB97XD, O3LYP and B3LYP functional with 6-31+G(d,p) basis set. It is observed from our studies that most of the descriptors presented show association with some processes, including absorption, blood-brain barrier transport, binding and even toxicity. Hence, the treatment of COVID-19 using Hydroxychloroquine and Azithromycin in some patients as single dose and their combination in patients with Corona virus resistance can be more effective. Our results show that these therapeutic molecules may also have good nonlinear optical applications, may have semiconductor character with wide band gap and can also be promising materials in the production of optoelectronic devices. The density of states and thermodynamic properties were equally determined.

In the present study, a systematic effort has been made to predict the hemolytic potency of chemically modified peptides. All models have been trained, tested, and evaluated on a dataset that contains 583 modified hemolytic peptides and a balanced number of non-hemolytic peptides. Machine learning techniques have been used to build the classification models using an immense range of peptide features that include 2D, 3D descriptors, fingerprints, atom, and diatom compositions. Random Forest based model developed using fingerprints as an input feature achieved maximum accuracy of 78.33% with AUC of 0.86 on the main dataset and accuracy of 78.29% with AUC of 0.85 on the validation dataset. Models developed in this study have been incorporated in a web server “HemoPImod” to facilitate the scientific community (http://webs.iiitd.edu.in/raghava/hemopimod/).

In the present study, a systematic effort has been made to predict the hemolytic potency of chemically modified peptides. All models have been trained, tested, and evaluated on a dataset that contains 583 modified hemolytic peptides and a balanced number of non-hemolytic peptides. Machine learning techniques have been used to build the classification models using an immense range of peptide features that include 2D, 3D descriptors, fingerprints, atom, and diatom compositions. Random Forest based model developed using fingerprints as an input feature achieved maximum accuracy of 78.33% with AUC of 0.86 on the main dataset and accuracy of 78.29% with AUC of 0.85 on the validation dataset. Models developed in this study have been incorporated in a web server "HemoPImod" to facilitate the scientific community (http://webs.iiitd.edu.in/raghava/hemopimod/). [ABSTRACT FROM AUTHOR]

The computational prediction of ligand-biopolymer affinities is a crucial endeavor in modern drug discovery and one that still poses major challenges. The choice of the appropriate computational method often reveals itself as a trade-off between accuracy and speed, with mathematical devices referred to as scoring functions being the fastest. Among the many shortcomings of scoring functions there is the lack of universal applicability to every molecular system. This is so largely due to their reliance on atom type perception and/or parametrization. This article proposes the use of nonparametric Model of Effective Radii of Atoms descriptors that can be readily computed for the entire Periodic Table and demonstrate that, in combination with machine learning algorithms, they can yield competitive performances and chemically meaningful insights. [ABSTRACT FROM AUTHOR]

Pseudomonas aeruginosa is a Gram-negative bacillus included among the six “ESKAPE” microbial species with an outstanding ability to “escape” currently used antibiotics and developing new antibiotics against it is of the highest priority. Whereas minimum inhibitory concentration (MIC) values against Pseudomonas aeruginosa have been used previously for QSAR model development, disk diffusion results (inhibition zones) have not been apparently used for this purpose in the literature and we decided to explore their use in this sense. We developed multiple QSAR methods using several machine learning algorithms (support vector classifier, K nearest neighbors, random forest classifier, decision tree classifier, AdaBoost classifier, logistic regression and naïve Bayes classifier). We used four sets of molecular descriptors and fingerprints and three different methods of data balancing, together with the “native” data set. In total, 32 models were built for each set of descriptors or fingerprint and balancing method, of which 28 were selected and stacked to create meta-models. In terms of balanced accuracy, the best performance was provided by KNN, logistic regression and decision tree classifier, but the ensemble method had slightly superior results in nested cross-validation.

The rising atmospheric CO2 level is partly responsible for global warming. Despite numerous warnings from scientists during the past years, nations are reacting too slowly, and thus, we will probably reach a situation needing rapid and effective techniques to reduce atmospheric CO2. Therefore, advanced engineering methods are particularly important to decrease the greenhouse effect, for instance, by capturing CO2 using solvents. Experimental testing of many solvents under different conditions is necessary but time-consuming. Alternatively, modeling CO2 capture by solvents using a nonlinear fitting machine learning is a rapid way to select potential solvents, prior to experimentation. Previous predictive machine learning models were mainly designed for blended solutions in water using the solution concentration as the main input of the model, which was not able to predict CO2 solubility in different types of physical solvents. To address this issue, here, we developed a new descriptor-based chemoinformatics model for predicting CO2 solubility in physical solvents in the form of mole fraction. The input factors include organic structural and bond information, thermodynamic properties, and experimental conditions. We studied the solvents from 823 data, including methanol (165 data), ethanol (138), n-propanol (98), n-butanol (64), n-pentanol (59), ethylene glycol (52), propylene glycol (54), acetone (51), 2-butanone (49), ethylene glycol monomethyl ether (46 data), and ethylene glycol monoethyl ether (47), using artificial neural networks as the machine learning model. Results show that our descriptor-based model predicts the CO2 absorption in physical solvents with generally higher accuracy and low root-mean-squared errors. Our findings show that using a set of simple but effective chemoinformatics-based descriptors, intrinsic relationships between the general properties of physical solvents and their CO2 solubility can be precisely fitted with machine learning. [ABSTRACT FROM AUTHOR]

MXenes are two-dimensional nanomaterials isolated from MAX phases by selective extraction of the A component—a p-block element. The MAX exfoliation energy, Eexf, is considered a chemical descriptor of the MXene synthesizability. Here, we show, by density functional theory (DFT) estimations of Eexf values for 486 different MAX phases, that Eexf decreases (i) when MAX is a nitride, (ii) when going along a metal M component d series, (iii) when going down a p-block A element group, and (iv) when having thicker MXenes. Furthermore, Eexf is found to bias, even to govern, the surface chemical activity, evaluated here on the CO2 adsorption strength, so that more unstable MXenes, displaying larger Eexf values, display a stronger attachment of species upon.

Currently, there are more than 1.5 million knee and hip replacement procedures carried out in the United States. Implants have a 10−15-year lifespan with up to 30% of revision surgeries showing complications with osteomyelitis. Titanium and titanium alloys are the favored implant materials because they are lightweight and have high mechanical strength. However, this increased strength can be associated with decreased bone density around the implant, leading to implant loosening and failure. To avoid this, current strategies include plasma-spraying titanium surfaces and foaming titanium. Both techniques give the titanium a rough and irregular finish that improves biocompatibility. Recently, researchers have also sought to surface-conjugate proteins to titanium to induce osteointegration. Cell adhesion-promoting proteins can be conjugated to methacrylate groups and crosslinked using a variety of methods. Methacrylated proteins can be conjugated to titanium surfaces through atom transfer radical polymerization (ATRP). However, surface conjugation of proteins increases biocompatibility non-specifically to bone cells, adding to the risk of biofouling which may result in osteomyelitis that causes implant failure. In this work, we analyze the factors contributing to biofouling when coating titanium to improve biocompatibility, and design an experimental scheme to evaluate optimal coating parameters.

Designing drug delivery vehicles using cell-penetrating peptides is a hot area of research in the field of medicine. In the past, number of in silico methods have been developed for predicting cell-penetrating property of peptides containing natural residues. In this study, first time attempt has been made to predict cell-penetrating property of peptides containing natural and modified residues. The dataset used to develop prediction models, include structure and sequence of 732 chemically modified cell-penetrating peptides and an equal number of non-cell penetrating peptides. We analyzed the structure of both class of peptides and observed that positive charge groups, atoms, and residues are preferred in cell-penetrating peptides. In this study, models were developed to predict cell-penetrating peptides from its tertiary structure using a wide range of descriptors (2D, 3D descriptors, and fingerprints). Random Forest model developed by using PaDEL descriptors (combination of 2D, 3D, and fingerprints) achieved maximum accuracy of 95.10%, MCC of 0.90 and AUROC of 0.99 on the main dataset. The performance of model was also evaluated on validation/independent dataset which achieved AUROC of 0.98. In order to assist the scientific community, we have developed a web server “CellPPDMod” for predicting the cell-penetrating property of modified peptides (http://webs.iiitd.edu.in/raghava/cellppdmod/).

Chemical descriptors using DFT concepts characterize elements reactivity. Hardness and electrophilicity relate to the charge and energy exchanged, determining changes in the length and slope during reaction. Three zones are identified. At low charge exchange (<0.25) the length of the path shortens with constant slope. At higher charge exchange (>0.50) the length increases while its slope rotates. The two regions respectively correspond to frontier molecular orbital and charge-controlled reactions. They relate to the divergence and rotational components of the chemical potential gradient, pointing toward nucleophilic or electrophilic reactions. In between reactions are a mixing both end-member components. [ABSTRACT FROM AUTHOR]

Designing drug delivery vehicles using cell-penetrating peptides is a hot area of research in the field of medicine. In the past, number of in silico methods have been developed for predicting cell-penetrating property of peptides containing natural residues. In this study, first time attempt has been made to predict cell-penetrating property of peptides containing natural andmodified residues. The dataset used to develop predictionmodels, include structure and sequence of 732 chemicallymodified cell-penetrating peptides and an equal number of non-cell penetrating peptides. We analyzed the structure of both class of peptides and observed that positive charge groups, atoms, and residues are preferred in cell-penetrating peptides. In this study, models were developed to predict cell-penetrating peptides from its tertiary structure using a wide range of descriptors (2D, 3D descriptors, and fingerprints). Random Forest model developed by using PaDEL descriptors (combination of 2D, 3D, and fingerprints) achieved maximum accuracy of 95.10%, MCC of 0.90 and AUROC of 0.99 on the main dataset. The performance of model was also evaluated on validation/independent dataset which achieved AUROC of 0.98. In order to assist the scientific community, we have developed a web server “CellPPDMod” for predicting the cell-penetrating property of modified peptides (http:// webs.iiitd.edu.in/raghava/cellppdmod/). [ABSTRACT FROM AUTHOR]

Nature plays a major role in the development of new drugs which helps in preventing and treating human diseases. Anti-obesity compound database (AOCD) contains comprehensive information on all published small molecules from natural sources with anti-obesity potential targeting pancreatic lipase (PL), appetite suppressant (AS) and adipogenesis (AD). Presently the database contains 349 compounds isolated from 307 plants, 26 marine and 16 microbial sources. Users can query the AOCD database (https://aocd.swmd.co.in/) in several ways. The database was divided into three datasets (PL, AS and AD) to perform chemoinformatic analysis using Platform for Unified Molecular Analysis (PUMA), which were analyzed based on molecular descriptors, scaffold diversity and structural fingerprint diversity. Chemoinformatics study inferred the PL dataset has the highest diversity of compounds based on the Euclidean distance on molecular properties, scaffold diversity and pairwise similarity on fingerprint diversity. This study would hasten the process of anti-obesity drug discovery. [Display omitted] • A comprehensive cheminformatics study of anti-obesity database (AOCD) based on their mechanism of action. • Pancreatic lipase (PL), Appetite suppressant (AS) and Adipocyte differentiation (AD) were the three datasets with 219, 91 and 101 compounds respectively. • Chemical space, scaffold diversity, and structural fingerprints were computed and analysed. • PL dataset in AOCD showed highest scaffold diversity. • Compounds diversity in AOCD includes flavonoids, terpenes, hydrolases, prenol lipids and polykides. [ABSTRACT FROM AUTHOR]

Previous modelling of the median lethal dose (oral rat LD50) has indicated that local class-based models yield better correlations than global models. We evaluated the hypothesis that dividing the dataset by pesticidal mechanisms would improve prediction accuracy. A linear discriminant analysis (LDA) based-approach was utilized to assign indicators such as the pesticide target species, mode of action, or target species - mode of action combination. LDA models were able to predict these indicators with about 87% accuracy. Toxicity is predicted utilizing the QSAR model fit to chemicals with that indicator. Toxicity was also predicted using a global hierarchical clustering (HC) approach which divides data set into clusters based on molecular similarity. At a comparable prediction coverage (~94%), the global HC method yielded slightly higher prediction accuracy (r2= 0.50) than the LDA method (r2~ 0.47). A single model fit to the entire training set yielded the poorest results (r2= 0.38), indicating that there is an advantage to clustering the dataset to predict acute toxicity. Finally, this study shows that whilst dividing the training set into subsets (i.e. clusters) improves prediction accuracy, it may not matter which method (expert based or purely machine learning) is used to divide the dataset into subsets. [ABSTRACT FROM PUBLISHER]

The mode of toxic action (MoA) has been recognized as a key determinant of chemical toxicity, but development of predictive MoA classification models in aquatic toxicology has been limited. We developed a Bayesian network model to classify aquatic toxicity MoA using a recently published dataset containing over one thousand chemicals with MoA assignments for aquatic animal toxicity. Two dimensional theoretical chemical descriptors were generated for each chemical using the Toxicity Estimation Software Tool. The model was developed through augmented Markov blanket discovery from the dataset of 1098 chemicals with the MoA broad classifications as a target node. From cross validation, the overall precision for the model was 80.2%. The best precision was for the AChEI MoA (93.5%) where 257 chemicals out of 275 were correctly classified. Model precision was poorest for the reactivity MoA (48.5%) where 48 out of 99 reactive chemicals were correctly classified. Narcosis represented the largest class within the MoA dataset and had a precision and reliability of 80.0%, reflecting the global precision across all of the MoAs. False negatives for narcosis most often fell into electron transport inhibition, neurotoxicity or reactivity MoAs. False negatives for all other MoAs were most often narcosis. A probabilistic sensitivity analysis was undertaken for each MoA to examine the sensitivity to individual and multiple descriptor findings. The results show that the Markov blanket of a structurally complex dataset can simplify analysis and interpretation by identifying a subset of the key chemical descriptors associated with broad aquatic toxicity MoAs, and by providing a computational chemistry-based network classification model with reasonable prediction accuracy. [ABSTRACT FROM AUTHOR]

Natural products play a significant role in cancer chemotherapy. They are likely to provide many lead structures, which can be used as templates for the construction of novel drugs with enhanced antitumor activity. Traditional research approaches studied structure-activity relationship of natural products and obtained key structural properties, such as chemical bond or group, with the purpose of ascertaining their effect on a single cell line or a single tissue type. Here, for the first time, we develop a machine learning method to comprehensively predict natural products responses against a panel of cancer cell lines based on both the gene expression and the chemical properties of natural products. The results on two datasets, training set and independent test set, show that this proposed method yields significantly better prediction accuracy. In addition, we also demonstrate the predictive power of our proposed method by modeling the cancer cell sensitivity to two natural products, Curcumin and Resveratrol, which indicate that our method can effectively predict the response of cancer cell lines to these two natural products. Taken together, the method will facilitate the identification of natural products as cancer therapies and the development of precision medicine by linking the features of patient genomes to natural product sensitivity.

The activity of a selected class of DPP4 inhibitors was preliminarily assessed using chemical descriptors derived AM1 optimized geometries. Using multiple linear regression model, it was found that ?E0, LUMO energy, area, molecular weight and ?H0 are the significant descriptors that can adequately assess the binding affinity of the compounds. The derived multiple linear regression (MLR) model was validated using rigorous statistical analysis. The preliminary model suggests that bulky and electrophilic inhibitors are desired.