Development of a robust and validated 2D-QSPR model for sweetness potency of diverse functional organic molecules

3D-QSAR studies on the structure-bitterness analysis of citrus flavonoids

Despite their important bioactivities, the unpleasant bitter taste of citrus derived flavonoids limits their applications in the food industry, and the structure-bitterness relationship of flavonoids is still far from clear. In this study, 26 flavonoids were characterized by their bitterness threshold and their common skeleton using sensory evaluation and molecular superposition, respectively. The quantitative conformational relationship of the structure-bitterness of flavonoids was explored using 3D-QSAR based on comparative molecular field analysis (CoMFA) and comparative molecular similarity index analysis (CoMSIA). The results showed that increases of a hydrogen bond donor at A-5 or B-3', a bulky group at A-8, or an electron-withdrawing group at B-4' would enhance the bitterness of flavonoids. The bitterness of some flavonoids was predicted and evaluated, and the results were similar to the bitter intensity of the counterparts from the 3D-QSAR and contour plots, confirming the validation of 3D-QSAR. This study explains the theory of the structure-bitterness relationship of flavonoids, by showing potential information for understanding the bitterness in citrus flavonoids and developing a debittering process.

Machine learning models to predict sweetness of molecules

Sweetness is a vital taste to which humans are innately attracted. Given the increasing prevalence of type-2 diabetes, it is highly relevant to build computational models to predict the sweetness of small molecules. Such models are valuable for identifying sweeteners with low calorific value. We present regression-based machine learning and deep learning algorithms for predicting sweetness. Toward this goal, we manually curated the most extensive dataset of 671 sweet molecules with known experimental sweetness values ranging from 0.2 to 22,500,000. Gradient Boost and Random Forest Regressors emerged as the best models for predicting the sweetness of molecules with a correlation coefficient of 0.94 and 0.92, respectively. Our models show state-of-the-art performance when compared with previously published studies. Besides making our dataset (SweetpredDB) available, we also present a user-friendly web server to return the predicted sweetness for small molecules, Sweetpred (https://cosylab.iiitd.edu.in/sweetpred).

Versatile in silico modelling of microplastics adsorption capacity in aqueous environment based on molecular descriptor and machine learning

To comprehensively evaluate the hazards of microplastics and their coexisting organic pollutants, the sorption capacity of microplastics is a major issue that is quantified through the microplastic-aqueous sorption coefficient (K_d). Almost all quantitative structure-property relationship (QSPR) models that describe K_d apply only to narrow, relatively homogeneous groups of reactants. Herein, non-hybrid QSPR-based models were developed to predict PE-water (K_PE-w), PE-seawater (K_PE-sw), PVC-water (K_PVC-w) and PP-seawater (K_PP-sw) sorption coefficients at different temperatures, with eight machine learning algorithms. Moreover, novel hybrid intelligent models for predicting K_d more accurately were innovatively developed by applying GA, PSO and AdaBoost algorithms to optimize MLP and ELM models. The results indicated that all three optimization algorithms could improve the robustness and predictability of the standalone MLP and ELM models. In all models trained with K_PE-w, K_PE-sw, K_PVC-w and K_PP-sw data sets, GBDT-1 and XGBoost-1 models, MLP-GA-2 and MLP-PSO-2 models, MLR-3 and MLR-4 models performed better in terms of goodness of fit (R_adj²: 0.907-0.999), robustness (Q_BOOT²: 0.900-0.937) and predictability (R_ext²: 0.889-0.970), respectively. Analyzing the descriptors revealed that temperature, lipophilicity, ionization potential and molecular size were correlated closely with the adsorption capacity of microplastics to organic pollutants. The proposed QSPR models may assist in initial environmental exposure assessments without imposing heavy costs in the early experimental phase.

A novel multi-layer prediction approach for sweetness evaluation based on systematic machine learning modeling

Nowadays, computational approaches have drawn more and more attention when exploring the relationship between sweetness and chemical structure instead of traditional experimental tests. In this work, we proposed a novel multi-layer sweetness evaluation system based on machine learning methods. It can be used to evaluate sweet properties of compounds with different chemical spaces and categories, including natural, artificial, carbohydrate, non-carbohydrate, nutritive and non-nutritive ones, suitable for different application scenarios. Furthermore, it provided quantitative predictions of sweetness. In addition, sweetness-related chemical basis and structure transforming rules were obtained by using molecular cloud and matched molecular pair analysis (MMPA) methods. This work systematically improved the data quality, explored the best machine learning algorithm and molecular characterizing strategy, and finally obtained robust models to establish a multi-layer prediction system (available at: https://github.com/ifyoungnet/ChemSweet). We hope that this study could facilitate food scientists with efficient screening and precise development of high-quality sweeteners.

Prediction reliability of QSAR models: an overview of various validation tools

The reliability of any quantitative structure-activity relationship (QSAR) model depends on multiple aspects such as the accuracy of the input dataset, selection of significant descriptors, the appropriate splitting process of the dataset, statistical tools used, and most notably on the measures of validation. Validation, the most crucial step in QSAR model development, confirms the reliability of the developed QSAR models and the acceptability of each step in the model development. The present review deals with various validation tools that involve multiple techniques that improve the model quality and robustness. The double cross-validation tool helps in building improved quality models using different combinations of the same training set in an inner cross-validation loop. This exhaustive method is also integrated for small datasets (< 40 compounds) in another tool, namely the small dataset modeler tool. The main aim of QSAR researchers is to improve prediction quality by lowering the prediction errors for the query compounds. 'Intelligent' selection of multiple models and consensus predictions integrated in the intelligent consensus predictor tool were found to be more externally predictive than individual models. Furthermore, another tool called Prediction Reliability Indicator was explained to understand the quality of predictions for a true external set. This tool uses a composite scoring technique to identify query compounds as 'good' or 'moderate' or 'bad' predictions. We have also discussed a quantitative read-across tool which predicts a chemical response based on the similarity with structural analogues. The discussed tools are freely available from https://dtclab.webs.com/software-tools or http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ and https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home (for read-across).

Analysis of Influencing Factors on the Gas Separation Performance of Carbon Molecular Sieve Membrane Using Machine Learning Technique

Gas separation performance of the carbon molecular sieve (CMS) membrane is influenced by multiple factors including the microstructural characteristics of carbon and gas properties. In this work, the support vector regression (SVR) method as a machine learning technique was applied to the correlation between the gas separation performance, the multiple membrane structure, and gas characteristic factors of the self-manufactured CMS membrane. A simple quantitative index based on the Robeson’s upper bound line, which indicated the gas permeability and selectivity simultaneously, was proposed to measure the gas separation performance of CMS membrane. Based on the calculation results, the inferred key factors affecting the gas permeability of CMS membrane were the fractional free volume (FFV) of the precursor, the average interlayer spacing of graphite-like carbon sheet, and the final carbonization temperature. Moreover, the most influential factors for the gas separation performance were supposed to be the two structural factors of precursor influencing the porosity of CMS membrane, the carbon residue and the FFV, and the ratio of the gas kinetic diameters. The results would be helpful to the structural optimization and the separation performance improvement of CMS membrane.

Exploring biological efficacy of novel benzothiazole linked 2,5-disubstituted-1,3,4-oxadiazole hybrids as efficient α-amylase inhibitors: Synthesis, characterization, inhibition, molecular docking, molecular dynamics and Monte Carlo based QSAR studies

In an effort to explore a class of novel antidiabetic agents, we have made an effort to synergize the α-amylase inhibitory potential of 1,3-benzothiazole and 1,3,4-oxadiazole scaffolds by combining the two into a single structure via an ether linkage. The structure of synthesized benzothiazole clubbed oxadiazole derivatives are established by different spectral techniques. The synthesized hybrids are evaluated for their in vitro inhibitory potential against α-amylase. Compound 8f is found to be the most potent with a significant inhibition (87.5 ± 0.74% at 50 μg/mL, 82.27 ± 1.85% at 25 μg/mL and 79.94 ± 1.88% at 12.5 μg/mL) when compared to positive control acarbose (77.96 ± 2.06%, 71.17 ± 0.60%, 67.24 ± 1.16% at 50 μg/mL, 25 μg/mL and 12.5 μg/mL concentration). Molecular docking of the most potent enzyme inhibitor, 8f, shows promising interaction with the binding site of biological macromolecule Aspergillus oryzae α-amylase (PDB ID: 7TAA) and human pancreatic α-amylase (PDB ID: 3BAJ). To a step further, in-depth QSAR studies show a significant correlation between the experimental and the predicted inhibitory activities with the best R_validation²= 0.8701. The developed QSAR model can provide ample information about the structural features responsible for the increase and decrease of inhibitory activity. The mechanistic interpretation of the structure-activity relationship (SAR) is done with the help of combined computational calculations i.e. molecular docking and QSAR. Finally, molecular dynamic simulations are performed to get an insight into the binding mode of the most potent derivative with α-amylase from A. oryzae (PDB ID: 7TAA) and human pancreas (PDB ID: 3BAJ).

Quantitative structure activity relationship studies of novel hydrazone derivatives as α-amylase inhibitors with index of ideality of correlation

The present manuscript describes the synthesis, α-amylase inhibition, in silico studies and in-depth quantitative structure-activity relationship (QSAR) of a library of aroyl hydrazones based on benzothiazole skeleton. All the compounds of the developed library are characterized by various spectral techniques. α-Amylase inhibitory potential of all compounds has been explored, where compound 7n exhibits remarkable α-amylase inhibition of 87.5% at 50 µg/mL. Robust QSAR models are made by using the balance of correlation method in CORAL software. The chemical structures at different concentration with optimal descriptors are represented by SMILES. A data set of 66 SMILES of 22 hydrazones at three distinct concentrations are prepared. The significance of the index of ideality of correlation (IIC) with applicability domain (AD) is also studied at depth. A QSAR model with best Rvalidation2 = 0.8587 for split 1 is considered as a leading model. The outliers and promoters of increase and decrease of endpoint are also extracted. The binding modes of the most active compound, that is, 7n in the active site of Aspergillus oryzae α-amylase (PDB ID: 7TAA) are also explored by in silico molecular docking studies. Compound 7n displays high resemblance in binding mode and pose with the standard drug acarbose. Molecular dynamics simulations performed on protein-ligand complex for 100 ns, the protein gets stabilised after 20 ns and remained below 2 Å for the remaining simulation. Moreover, the deviation observed in RMSF during simulation for each amino acid residue with respect to Cα carbon atom is insignificant.

QSAR modeling of algal low level toxicity values of different phenol and aniline derivatives using 2D descriptors

The deposition of different types of phenol and aniline derivatives in the aquatic environment leads to toxicity to living organisms. Under such condition, evaluation of these toxicants is very much important. Due to non-availability of sufficient experimental data as well as sufficient number of Quantitative Structure-Activity Relationship (QSAR) models for the low level toxicity values for such pollutants, we have employed here the partial least squares (PLS) regression for the development of robust and predictive QSAR models using low level toxicity values against algal species. Here, we have used both Extended Topochemical Atom (ETA) and non-ETA indices as 2D descriptors for model development. The statistical validation parameters ensure the robustness and the predictivity of the developed models. From the insights of the final PLS models, it can be concluded that presence of nitro groups (in the ortho position to phenolic hydroxyl group increasing intramolecular hydrogen bonding capacity), presence of chlorine substituents (influencing lipophilicity) especially at the para position, oxygen and nitrogen at the topological distance three, aliphatic side chain (contributing to hydrophobicity), molecules with large size atoms and higher molecular bulk will increase the toxicity towards the algal species. On the other hand, the phenol ring without any substituent or with a polar substituent (like amino group), presence of chlorine at ortho-ortho or ortho-para position, absence of nitro group, presence of chlorine and oxygen at the topological distance three, presence of lower number of aliphatic groups will decrease the toxic effect towards the algal species.

Structure-dependent effects of sweet and sweet taste affecting compounds on their sensorial properties

A reduction in sugar consumption is desirable from a health point of view. However, the sensory profiles of alternative sweet tasting compounds differ from sucrose regarding their temporal profile and undesired side tastes, reducing consumers' acceptance. The present study describes a sensory characterization of a variety of sweet and sweet taste affecting compounds followed by a comparison of similarity to sucrose and a multivariate regression analysis to investigate structural determinants and possible interactions for the temporal profile of the sweetness and side-tastes. The results of the present study suggest a pivotal role for the number of ketones, aromatic rings, double bonds and the M LogP in the temporal profile of sweet and sweet taste affecting compounds. Furthermore, interactions between aggregated physicochemical descriptors demonstrate the complexity of the sensory response, which should be considered in future models to predict a comprehensive sensory profile of sweet and sweet taste affecting compounds.

First report on a classification-based QSAR model for chemical toxicity to earthworm

As the use of the pesticides has increased extensively in the farming fields to have a better agricultural production, the negative impacts of such use have also increased exponentially. Hence, the toxic effects of pesticides along with the targeted organisms affect the non-targeted terrestrial organisms such as earthworm. Therefore, in the present work, we have developed a classification-based quantitative structure-activity relationship (QSAR) model using linear discriminant analysis (LDA) to capture the specific information of pesticides / diverse chemicals in order to determine the structural information responsible for toxicity manifestation towards the non-targeted organism, i.e., earthworm (Eisenia foetida). After variable selection, the model was developed using 2D descriptors only and was subjected to rigorous statistical validation. The best discriminant model obtained with 8 descriptors showed appreciable Wilks' λ value of 0.490, F (Fischer's statistics) value of 14.03, χ² value of 79.098, canonical regression coefficient (R) value of 0.714 and ρ value of 14.63. The sensitivity, specificity, accuracy, precision and F-measure values of the training set are 90.00, 80.52, 83.76, 70.59 and 79.12 respectively whereas for the test set, these are 58.82, 79.31, 71.74, 62.50 and 60.61 respectively. The insights obtained from the LDA model suggested that lipophilicity, electronrichness, and lower degree of branching of the organic compounds are responsible for earthworm toxicity through various mechanisms. On the other hand, polar and bulky diverse chemicals do not have such toxic effects on earthworm. Hence, this model can be an effective tool to tailor molecular structures of the existing pesticides to develop novel compounds or pesticides which would be less toxic to the non-targeted organisms, specifically earthworm.

Exploring indole derivatives as myeloid cell leukaemia-1 (Mcl-1) inhibitors with multi-QSAR approach: a novel hope in anti-cancer drug discovery

In humans, the over-expression of Mcl-1 protein causes different cancers and it is also responsible for cancer resistance to different cytotoxic agents.

Nucleobase sequence based building up of reliable QSAR models with the index of ideality correlation using Monte Carlo method

This study describes in silico designing of aptamers against the influenza virus using Monte Carlo method. Aptamers are short, single-stranded oligonucleotides and these bind to an ample range of biologically important proteins which are related to many disease conditions. The affinities and specificities of aptamers are comparable to antibodies. In the medicinal chemistry, quantitative structure-activity relationship (QSAR) is an important skill which is used for drug design and development. To study the inhibitory activity of aptamers, we have developed QSAR models based on Monte Carlo method. The nucleobase sequence descriptors B_k, BB_k and BBB_k are used to generate the QSAR models. A number of statistical benchmarks together with index of ideality of correlation (IIC) is considered to validate the build QSAR models. Data set of 98 aptamers is divided into four random splits. The statistical criteria R² = 0.8711 and CCC = 0.9207 of the validation set of split 3 are best, so the build QSAR model of split 3 is the paramount model. The aptamer fragment responsible for the promotors of endpoint increase and decrease are also determined. These fragments are applied to design new nine aptamers from the lead aptamer APT01.Communicated by Ramaswamy H. Sarma.

Exploring QSPR modeling for adsorption of hazardous synthetic organic chemicals (SOCs) by SWCNTs

In order to understand the physicochemical properties as well as the mechanisms behind adsorption of hazardous synthetic organic chemicals (SOCs) onto single walled carbon nanotubes (SWCNTs), we have developed partial least squares (PLS)-regression based QSPR models using a diverse set of 40 hazardous SOCs having defined adsorption coefficient (logK). The models were extensively validated using different validation parameters in order to assure the robustness and predictivity of the models. We have also checked the consensus predictivity of all the individual models using "Intelligent consensus predictor" tool for possible enhancement of the quality of predictions for test set compounds. The consensus predictivity of the test set compounds were found to be better than the individual models based on not only the MAE based criteria (MAE_(95%) = Good) but also some other validation parameters (Q²_F1 = 0.938, Q²_F2 = 0.937). The contributing descriptors obtained from the QSPR models suggested that the hazardous SOCs may get adsorbed onto the SWCNTs through hydrophobic interaction as well as hydrogen bonding interactions and electrostatic interaction to the functionally modified SWCNTs. Thus, the developed models may provide knowledge to scientists to increase the efficient application of SWCNTs as a special adsorbent, which may be useful for the management of environmental pollution.

The Acid Strength of the Lewis-Brønsted Superacids - A QSPR Study

The acidity of Lewis-Brønsted superacids can be derived from the theoretical calculations as the Gibbs free energy of the deprotonation reaction (ΔG_acid ), which describes the tendency of a studied compound to donate a proton. This paper presents the first Quantitative Structure - Property Relationship (QSPR) model that correlates the ΔG_acid of superacid (HF/MeX₃ formula (X=F, Cl, Br)) with their structure. Developed model is well fitted, roubustness, has good predictive abilities, fulfills all OECD recommendation for good model. Obtained results provide the insight into the relation of structural features of superacids, which are responsible for their acid strength - the structures characterized by strong F-Me dative bond (with relatively large vibrational frequency), small positive partial atomic charge on Me central atom, possibly large polarity exhibit large acid strength. Such assumption can be used in the future as valuable information in the process of the designing new, stronger, more effective superacids.

e-Sweet: A Machine-Learning Based Platform for the Prediction of Sweetener and Its Relative Sweetness

Artificial sweeteners (AS) can elicit the strong sweet sensation with the low or zero calorie, and are widely used to replace the nutritive sugar in the food and beverage industry. However, the safety issue of current AS is still controversial. Thus, it is imperative to develop more safe and potent AS. Due to the costly and laborious experimental-screening of AS, in-silico sweetener/sweetness prediction could provide a good avenue to identify the potential sweetener candidates before experiment. In this work, we curate the largest dataset of 530 sweeteners and 850 non-sweeteners, and collect the second largest dataset of 352 sweeteners with the relative sweetness (RS) from the literature. In light of these experimental datasets, we adopt five machine-learning methods and conformational-independent molecular fingerprints to derive the classification and regression models for the prediction of sweetener and its RS, respectively via the consensus strategy. Our best classification model achieves the 95% confidence intervals for the accuracy (0.91 ± 0.01), precision (0.90 ± 0.01), specificity (0.94 ± 0.01), sensitivity (0.86 ± 0.01), F1-score (0.88 ± 0.01), and NER (Non-error Rate: 0.90 ± 0.01) on the test set, which outperforms the model (NER = 0.85) of Rojas et al. in terms of NER, and our best regression model gives the 95% confidence intervals for the R²(test set) and ΔR² [referring to |R²(test set)- R²(cross-validation)|] of 0.77 ± 0.01 and 0.03 ± 0.01, respectively, which is also better than the other works based on the conformation-independent 2D descriptors (e.g., 2D Dragon) according to R²(test set) and ΔR². Our models are obtained by averaging over nineteen data-splitting schemes, and fully comply with the guidelines of Organization for Economic Cooperation and Development (OECD), which are not completely followed by the previous relevant works that are all on the basis of only one random data-splitting scheme for the cross-validation set and test set. Finally, we develop a user-friendly platform “e-Sweet” for the automatic prediction of sweetener and its corresponding RS. To our best knowledge, it is a first and free platform that can enable the experimental food scientists to exploit the current machine-learning methods to boost the discovery of more AS with the low or zero calorie content.

Prediction of the adsorption coefficients of some aromatic compounds on multi-wall carbon nanotubes by the Monte Carlo method

In this investigation, quantitative structure-property relationship (QSPR) modelling of adsorption coefficients of 69 aromatic compounds on multi-wall carbon nanotubes (MWCNTs) was studied using the Monte Carlo method. QSPR models were calculated with CORAL software, and optimal descriptors were calculated with the simplified molecular input line entry system (SMILES) and hydrogen-suppressed molecular graphs (HSGs). The aromatic compound data set was randomly split into training, invisible training, calibration and validation sets. Analysis of three probes of the Monte Carlo optimization with three random splits was done. The results from three random splits displayed robust, very simple, predictable and reliable models for the training, invisible training, calibration and validation sets with a coefficient of determination (r²) equal to 0.9463-0.8528, 0.9020-0.8324, 0.9606-0.9178 and 0.9573-0.8228, respectively. As a result, the models obtained help to identify the hybrid descriptors for the increase and the decrease of the adsorption coefficient of aromatic compounds on MWCNTs. This simple QSPR model can be used for the prediction of the adsorption coefficient of numerous aromatic compounds on MWCNTs.

How Precise Are Our Quantitative Structure-Activity Relationship Derived Predictions for New Query Chemicals?

Quantitative structure-activity relationship (QSAR) models have long been used for making predictions and data gap filling in diverse fields including medicinal chemistry, predictive toxicology, environmental fate modeling, materials science, agricultural science, nanoscience, food science, and so forth. Usually a QSAR model is developed based on chemical information of a properly designed training set and corresponding experimental response data while the model is validated using one or more test set(s) for which the experimental response data are available. However, it is interesting to estimate the reliability of predictions when the model is applied to a completely new data set (true external set) even when the new data points are within applicability domain (AD) of the developed model. In the present study, we have categorized the quality of predictions for the test set or true external set into three groups (good, moderate, and bad) based on absolute prediction errors. Then, we have used three criteria [(a) mean absolute error of leave-one-out predictions for 10 most close training compounds for each query molecule; (b) AD in terms of similarity based on the standardization approach; and (c) proximity of the predicted value of the query compound to the mean training response] in different weighting schemes for making a composite score of predictions. It was found that using the most frequently appearing weighting scheme 0.5-0-0.5, the composite score-based categorization showed concordance with absolute prediction error-based categorization for more than 80% test data points while working with 5 different datasets with 15 models for each set derived in three different splitting techniques. These observations were also confirmed with true external sets for another four endpoints suggesting applicability of the scheme to judge the reliability of predictions for new datasets. The scheme has been implemented in a tool "Prediction Reliability Indicator" available at http://dtclab.webs.com/software-tools and http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/, and the tool is presently valid for multiple linear regression models only.

Consensus Predictive Model for Human K562 Cell Growth Inhibition through Enalos Cloud Platform

β-Thalassemia is an inherited hematologic disorder caused by various mutations of the β-globin gene, thus resulting in a significant decrease in adult hemoglobin (HbA) production. An increase in fetal hemoglobin (HbF) levels by drug molecules is considered of great potential in β-thalassemia treatment and is expected to counterbalance the impaired production of HbA. In this work, based on a set of 129 experimentally tested biological inhibitors, we developed and validated a computational model for the prediction of K562 functional inhibition, possibly associated with HbF induction. To facilitate future advancements in the field, we incorporated our model into Enalos Cloud Platform, which enabled online access to our computational scheme (http://enalos.insilicotox.com/K562) through a user-friendly interface. This web service is offered to the wider community to promote in silico drug discovery through fast and reliable predictions.

Chemometric modeling of odor threshold property of diverse aroma components of wine

We have modelled here odor threshold properties (OTP) of various aroma components present in different types of wine using quantitative structure–property relationship (QSPR) studies employing both two-dimensional and three-dimensional descriptors.

Enalos+ KNIME Nodes: New Cheminformatics Tools for Drug Discovery

In this chapter we present and discuss Enalos+ nodes designed and developed by NovaMechanics Ltd. for the open-source KNIME platform, as a useful aid when dealing with cheminformatics and nanoinformatics problems or medicinal applications. Enalos+ nodes facilitate tasks performed in molecular modeling and allow access, data mining, and manipulation for multiple chemical databases through the KNIME interface. Enalos+ nodes automate common procedures that greatly facilitate the rapid workflow prototyping within KNIME. Μethods and techniques that are included in Enalos+ nodes are presented in order to offer a deeper understanding of the theoretical background of the incorporated functionalities. An emphasis is given to demonstrate the usefulness of Enalos+ nodes in different cheminformatics applications by presenting four indicative case studies. Specifically, we present case studies that underline the value and the effectiveness of the nodes for molecular descriptors calculation and QSAR predictive model development. In addition, case studies are also presented demonstrating the benefits of the use of Enalos+ nodes for database exploitation within a drug discovery project.

Enalos Suite: New Cheminformatics Platform for Drug Discovery and Computational Toxicology

In this chapter we present and discuss, with the aid of several representative case studies from drug discovery and computational toxicology, a new cheminformatics platform, Enalos Suite, that was developed with open source and freely available software. Enalos Suite ( http://enalossuite.novamechanics.com/ ) was designed and developed as a useful tool to address a variety of cheminformatics problems, given that it expedites tasks performed in predictive modeling and allows access, data mining and manipulation for multiple chemical databases (PubChem, UniChem, etc.). Enalos Suite was carefully designed to permit its extension and adjustment to the special field of interest of each user, including, for instance, nanoinformatics, biomedical, and other applications. To demonstrate the functionalities of Enalos Suite that are useful in different cheminformatics applications, we present indicative case studies that include the exploitation of chemical databases within a drug discovery project, the calculation of molecular descriptors, and finally the development of a predictive QSAR model validated according to OECD principles. We aspire that at the end of this chapter, the reader will capture the effectiveness of different functionalities included in the Enalos Suite that could be of significant value in a multitude of cheminformatics applications.

PLS regression-based chemometric modeling of odorant properties of diverse chemical constituents of black tea and coffee

We investigate the key structural features regulating the odorant properties of constituents present in black tea and coffee, the most attractive non-alcoholic beverages.

A QSTR-Based Expert System to Predict Sweetness of Molecules

This work describes a novel approach based on advanced molecular similarity to predict the sweetness of chemicals. The proposed Quantitative Structure-Taste Relationship (QSTR) model is an expert system developed keeping in mind the five principles defined by the Organization for Economic Co-operation and Development (OECD) for the validation of (Q)SARs. The 649 sweet and non-sweet molecules were described by both conformation-independent extended-connectivity fingerprints (ECFPs) and molecular descriptors. In particular, the molecular similarity in the ECFPs space showed a clear association with molecular taste and it was exploited for model development. Molecules laying in the subspaces where the taste assignation was more difficult were modeled trough a consensus between linear and local approaches (Partial Least Squares-Discriminant Analysis and N-nearest-neighbor classifier). The expert system, which was thoroughly validated through a Monte Carlo procedure and an external set, gave satisfactory results in comparison with the state-of-the-art models. Moreover, the QSTR model can be leveraged into a greater understanding of the relationship between molecular structure and sweetness, and into the design of novel sweeteners.