Consistency of QSAR models: Correct split of training and test sets, ranking of models and performance parameters

CSEL-BGC: A Bioinformatics Framework Integrating Machine Learning for Defining the Biosynthetic Evolutionary Landscape of Uncharacterized Antibacterial Natural Products

The sluggish pace of new antibacterial drug development reflects a vulnerability in the face of the current severe threat posed by bacterial resistance. Microbial natural products (NPs), as a reservoir of immense chemical potential, have emerged as the most promising avenue for the discovery of next generation antibacterial agent. Directly accessing the antibacterial activity of potential products derived from biosynthetic gene clusters (BGCs) would significantly expedite the process. To tackle this issue, we propose a CSEL-BGC framework that integrates machine learning (ML) techniques. This framework involves the development of a novel cascade-stacking ensemble learning (CSEL) model and the establishment of a groundbreaking model evaluation system. Based on this framework, we predict 6,666 BGCs with antibacterial activity from 3,468 complete bacterial genomes and elucidate a biosynthetic evolutionary landscape to reveal their antibacterial potential. This provides crucial insights for interpretating the synthesis and secretion mechanisms of unknown NPs.

Design of vilazodone-donepezil chimeric derivatives as acetylcholinesterase inhibitors by QSAR, molecular docking and molecular dynamics simulations

Alzheimer's disease (AD) is a disease that affects the cognitive abilities of older adults, and it is one of the biggest global medical challenges of the 21st century. Acetylcholinesterase (AChE) can increase acetylcholine concentrations and improve cognitive function in patients, and is a potential target to develop small molecule inhibitors for the treatment of Alzheimer's disease (AD). In this study, 29 vilazodone-donepezil chimeric derivatives are systematically studied using 3D-QSAR modeling, and a robust and reliable Topomer CoMFA model was obtained with: q2 = 0.720, r2 = 0.991, F = 287.234, N = 6, and SEE = 0.098. Based on the established model and combined with the ZINC20 database, 33 new compounds with ideal inhibitory activity are successfully designed. Molecular docking and ADMET property prediction also show that these newly designed compounds have a good binding ability to the target protein and can meet the medicinal conditions. Subsequently, four new compounds with good comprehensive ability are selected for molecular dynamics simulation, and the simulation results confirm that the newly designed compounds have a certain degree of reliability and stability. This study provides guidance for vilazodone-donepezil chimeric derivatives as a potential AChE inhibitor and has certain theoretical value.

QSAR Study, Molecular Docking and Molecular Dynamic Simulation of Aurora Kinase Inhibitors Derived from Imidazo[4,5-b]pyridine Derivatives

Cancer is a serious threat to human life and social development and the use of scientific methods for cancer prevention and control is necessary. In this study, HQSAR, CoMFA, CoMSIA and TopomerCoMFA methods are used to establish models of 65 imidazo[4,5-b]pyridine derivatives to explore the quantitative structure-activity relationship between their anticancer activities and molecular conformations. The results show that the cross-validation coefficients q2 of HQSAR, CoMFA, CoMSIA and TopomerCoMFA are 0.892, 0.866, 0.877 and 0.905, respectively. The non-cross-validation coefficients r2 are 0.948, 0.983, 0.995 and 0.971, respectively. The externally validated complex correlation coefficients r2pred of external validation are 0.814, 0.829, 0.758 and 0.855, respectively. The PLS analysis verifies that the QSAR models have the highest prediction ability and stability. Based on these statistics, virtual screening based on R group is performed using the ZINC database by the Topomer search technology. Finally, 10 new compounds with higher activity are designed with the screened new fragments. In order to explore the binding modes and targets between ligands and protein receptors, these newly designed compounds are conjugated with macromolecular protein (PDB ID: 1MQ4) by molecular docking technology. Furthermore, to study the nature of the newly designed compound in dynamic states and the stability of the protein-ligand complex, molecular dynamics simulation is carried out for N3, N4, N5 and N7 docked with 1MQ4 protease structure for 50 ns. A free energy landscape is computed to search for the most stable conformation. These results prove the efficient and stability of the newly designed compounds. Finally, ADMET is used to predict the pharmacology and toxicity of the 10 designed drug molecules.

Combining QSAR techniques, molecular docking, and molecular dynamics simulations to explore anti-tumor inhibitors targeting Focal Adhesion Kinase

Focal Adhesion Kinase (FAK) is an important target for tumor therapy and is closely related to tumor cell genesis and progression. In this paper, we selected 46 FAK inhibitors with anticancer activity in the pyrrolo pyrimidine backbone to establish 3D/2D-QSAR models to explore the relationship between inhibitory activity and molecular structure. We have established two ideal models, namely, the Topomer CoMFA model (q 2 = 0.715, r 2 = 0.984) and the Holographic Quantitative Structure-Activity Relationship (HQSAR) model (q 2 = 0.707, r 2 = 0.899). Both models demonstrate excellent external prediction capabilities.Based on the QSAR results, we designed 20 structurally modified novel compounds, which were subjected to molecular docking and molecular dynamics studies, and the results showed that the new compounds formed many robust interactions with residues within the active pocket and could maintain stable binding to the receptor proteins. This study not only provides a powerful screening tool for designing novel FAK inhibitors, but also presents a series of novel FAK inhibitors with high micromolar activity that can be used for further characterization. It provides a reference for addressing the shortcomings of drug metabolism and drug resistance of traditional FAK inhibitors, as well as the development of novel clinically applicable FAK inhibitors.Communicated by Ramaswamy H. Sarma.

Artificial intelligence: Machine learning approach for screening large database and drug discovery

Recent research in drug discovery dealing with many faces difficulties, including development of new drugs during disease outbreak and drug resistance due to rapidly accumulating mutations. Virtual screening is the most widely used method in computer aided drug discovery. It has a prominent ability in screening drug targets from large molecular databases. Recently, a number of web servers have developed for quickly screening publicly accessible chemical databases. In a nutshell, deep learning algorithms and artificial neural networks have modernised the field. Several drug discovery processes have used machine learning and deep learning algorithms, including peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modelling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Although there are presently a wide variety of data-driven AI/ML tools available, the majority of these tools have, up to this point, been developed in the context of non-communicable diseases like cancer, and a number of obstacles have prevented the translation of these tools to the discovery of treatments against infectious diseases. In this review various aspects of AI and ML in virtual screening of large databases were discussed. Here, with an emphasis on antivirals as well as other disease, offers a perspective on the advantages, drawbacks, and hazards of AI/ML techniques in the search for innovative treatments.

Chemical similarity of molecules with physiological response

Measuring the similarity among molecules is an important task in various chemically oriented problems. This elusive concept is hard to define and quantify. Moreover, the complexity of the problem is elevated by bifurcating the notion of molecular similarity to structural and chemical similarity. While the structural similarity of molecules is being extensively researched, the so-called chemical similarity is being mentioned scarcely. Here, we propose a way of converting the physico-chemical properties into molecular fingerprints. Then, using the apparatus of measuring the structural similarity, the chemical similarity can be assessed. The proof of a concept is demonstrated on a set of molecules that induce diverse physiological responses.

Identification of Coronary Artery Diseases Using Photoplethysmography Signals and Practical Feature Selection Process

A low-cost, fast, dependable, repeatable, non-invasive, portable, and simple-to-use vascular screening tool for coronary artery diseases (CADs) is preferred. Photoplethysmography (PPG), a low-cost optical pulse wave technology, is one method with this potential. PPG signals come from changes in the amount of blood in the microvascular bed of tissue. Therefore, these signals can be used to figure out anomalies within the cardiovascular system. This work shows how to use PPG signals and feature selection-based classifiers to identify cardiorespiratory disorders based on the extraction of time-domain features. Data were collected from 360 healthy and cardiovascular disease patients. For analysis and identification, five types of cardiovascular disorders were considered. The categories of cardiovascular diseases were identified using a two-stage classification process. The first stage was utilized to differentiate between healthy and unhealthy subjects. Subjects who were found to be abnormal were then entered into the second stage classifier, which was used to determine the type of the disease. Seven different classifiers were employed to classify the dataset. Based on the subset of features found by the classifier, the Naïve Bayes classifier obtained the best test accuracy, with 94.44% for the first stage and 89.37% for the second stage. The results of this study show how vital the PPG signal is. Many time-domain parts of the PPG signal can be easily extracted and analyzed to find out if there are problems with the heart. The results were accurate and precise enough that they did not need to be looked at or analyzed further. The PPG classifier built on a simple microcontroller will work better than more expensive ones and will not make the patient nervous.

The Relevance of Goodness-of-fit, Robustness and Prediction Validation Categories of OECD-QSAR Principles with Respect to Sample Size and Model Type

We investigated the relevance of the validation principles on the Quantitative Structure Activity Relationship models issued by Organization for Economic and Co-operation and Development. We checked the goodness-of-fit, robustness and predictivity categories in linear and nonlinear models using benchmark datasets. Most of our conclusions are drawn using the sample size dependence of the different validation parameters. We found that the goodness-of-fit parameters misleadingly overestimate the models on small samples. In the case of neural network and support vector models, the feasibility of the goodness-of-fit parameters often might be questioned. We propose to use the simplest y-scrambling method to estimate chance correlation. We found that the leave-one-out and leave-many-out cross-validation parameters can be rescaled to each other in all models and the computationally feasible method should be chosen depending on the model type. We assessed the interdependence of the validation parameters by calculating their rank correlations. Goodness of fit and robustness correlate quite well over a sample size for linear models and one of the approaches might be redundant. In the rank correlation between internal and external validation parameters, we found that the assignment of good and bad modellable data to the training or the test causes negative correlations.

Comparison of various methods for validity evaluation of QSAR models

BACKGROUND

Quantitative structure-activity relationship (QSAR) modeling is one of the most important computational tools employed in drug discovery and development. The external validation of QSAR models is the main point to check the reliability of developed models for the prediction activity of not yet synthesized compounds. It was performed by different criteria in the literature.

METHODS

In this study, 44 reported QSAR models for biologically active compounds reported in scientific papers were collected. Various statistical parameters of external validation of a QSAR model were calculated, and the results were discussed.

RESULTS

The findings revealed that employing the coefficient of determination (r²) alone could not indicate the validity of a QSAR model. The established criteria for external validation have some advantages and disadvantages which should be considered in QSAR studies.

CONCLUSION

This study showed that these methods alone are not only enough to indicate the validity/invalidity of a QSAR model.

Comparison of Descriptor- and Fingerprint Sets in Machine Learning Models for ADME-Tox Targets

The screening of compounds for ADME-Tox targets plays an important role in drug design. QSPR models can increase the speed of these specific tasks, although the performance of the models highly depends on several factors, such as the applied molecular descriptors. In this study, a detailed comparison of the most popular descriptor groups has been carried out for six main ADME-Tox classification targets: Ames mutagenicity, P-glycoprotein inhibition, hERG inhibition, hepatotoxicity, blood–brain-barrier permeability, and cytochrome P450 2C9 inhibition. The literature-based, medium-sized binary classification datasets (all above 1,000 molecules) were used for the model building by two common algorithms, XGBoost and the RPropMLP neural network. Five molecular representation sets were compared along with their joint applications: Morgan, Atompairs, and MACCS fingerprints, and the traditional 1D and 2D molecular descriptors, as well as 3D molecular descriptors, separately. The statistical evaluation of the model performances was based on 18 different performance parameters. Although all the developed models were close to the usual performance of QSPR models for each specific ADME-Tox target, the results clearly showed the superiority of the traditional 1D, 2D, and 3D descriptors in the case of the XGBoost algorithm. It is worth trying the classical tools in single model building because the use of 2D descriptors can produce even better models for almost every dataset than the combination of all the examined descriptor sets.

Multiobject Optimization of National Football League Drafts: Comparison of Teams and Experts

Predicting the success of National Football League drafts has always been an exciting issue for the teams, fans and even for scientists. Among the numerous approaches, one of the best techniques is to ask the opinion of sport experts, who have the knowledge and past experiences to rate the drafts of the teams. When asking a set of sport experts to evaluate the performances of teams, a multicriteria decision making problem arises unavoidably. The current paper uses the draft evaluations of the 32 NFL teams given by 18 experts: a novel multicriteria decision making tool has been applied: the sum of ranking differences (SRD). We introduce a quick and easy-to-follow approach on how to evaluate the performance of the teams and the experts at the same time. Our results on the 2021 NFL draft data indicate that Green Bay Packers has the most promising drafts for 2021, while the experts have been grouped into three distinct groups based on the distance to the hypothetical best evaluation. Even the coding options can be tailored according to the experts’ opinions. Statistically correct (pairwise or group) comparisons can be made using analysis of variance (ANOVA). A comparison to TOPSIS ranking revealed that SRD gives a more objective ranking due to the lack of predefined weights.

Factor analysis, sparse PCA, and Sum of Ranking Differences-based improvements of the Promethee-GAIA multicriteria decision support technique

The Promethee-GAIA method is a multicriteria decision support technique that defines the aggregated ranks of multiple criteria and visualizes them based on Principal Component Analysis (PCA). In the case of numerous criteria, the PCA biplot-based visualization do not perceive how a criterion influences the decision problem. The central question is how the Promethee-GAIA-based decision-making process can be improved to gain more interpretable results that reveal more characteristic inner relationships between the criteria. To improve the Promethee-GAIA method, we suggest three techniques that eliminate redundant criteria as well as clearly outline, which criterion belongs to which factor and explore the similarities between criteria. These methods are the following: A) Principal factoring with rotation and communality analysis (P-PFA), B) the integration of Sparse PCA into the Promethee II method (P-sPCA), and C) the Sum of Ranking Differences method (P-SRD). The suggested methods are presented through an I4.0+ dataset that measures the Industry 4.0 readiness of NUTS 2-classified regions. The proposed methods are useful tools for handling multicriteria ranking problems, if the number of criteria is numerous.

Comprehensible Visualization of Multidimensional Data: Sum of Ranking Differences-Based Parallel Coordinates

A novel visualization technique is proposed for the sum of ranking differences method (SRD) based on parallel coordinates. An axis is defined for each variable, on which the data are depicted row-wise. By connecting data, the lines may intersect. The fewer intersections between the variables, the more similar they are and the clearer the figure becomes. Therefore, the visualization depends on what techniques are used to order the variables. The key idea is to employ the SRD method to measure the degree of similarity of the variables, establishing a distance-based order. The distances between the axes are not uniformly distributed in the proposed visualization; their closeness reflects similarity, according to their SRD value. The proposed algorithm identifies false similarities through an iterative approach, where the angles between the SRD values determine which side a variable is plotted. Visualization of the algorithm is provided by MATLAB/Octave source codes. The proposed tool is applied to study how the sources of greenhouse gas emissions can be grouped based on the statistical data of the countries. A comparison to multidimensional scaling (MDS)-based ordering is also given. The use case demonstrates the applicability of the method and the synergies of the incorporation of the SRD method into parallel coordinates.

3D-QSAR Studies of 1,2,4-Oxadiazole Derivatives as Sortase A Inhibitors

Sortase A (SrtA) is an enzyme that catalyzes the attachment of proteins to the cell wall of Gram-positive bacterial membrane, preventing the spread of pathogenic bacterial strains. Here, one class of oxadiazole compounds was distinguished as an efficient inhibitor of SrtA via the "S. aureus Sortase A" substrate-based virtual screening. The current study on 3D-QSAR was done by utilizing preparation of the structure in the Schrödinger software suite and an assessment of 120 derivatives with the crystal structure of 1,2,4-oxadiazole which was extracted from the PDB data bank. The docking operation of the best compound in terms of pMIC (pMIC = 2.77) was done to determine the drug likeliness and binding form of 1,2,4-oxadiazole derivatives as antibiotics in the active site. Using the kNN-MFA way, seven models of 3D-QSAR were created and amongst them, and one model was selected as the best. The chosen model based on q 2 (pred_r 2) and R 2 values related to the sixth factor of PLS illustrates better and more acceptable external and internal predictions. Values of crossvalidation (pred_r 2), validation (q 2), and F were observed 0.5479, 0.6319, and 179.0, respectively, for a test group including 24 molecules and the training group including 96 molecules. The external reliability outcomes showed that the acceptable and the selective 3D-QSAR model had a high predictive potential (R 2 = 0.9235) which was confirmed by the Y-randomization test. Besides, the model applicability domain was described successfully to validate the estimation of the model.

Modelling in Synthesis and Optimization of Active Vaccinal Components

Cancer is the second leading cause of mortality worldwide, behind heart diseases, accounting for 10 million deaths each year. This study focusses on adenocarcinoma, which is a target of a number of anticancer therapies presently being tested in medical and pharmaceutical studies. The innovative study for a therapeutic vaccine comprises the investigation of gold nanoparticles and their influence on the immune response for the annihilation of cancer cells. The model is intended to be realized using Quantitative-Structure Activity Relationship (QSAR) methods, explicitly artificial neural networks combined with fuzzy rules, to enhance automated properties of neural nets with human perception characteristics. Image processing techniques such as morphological transformations and watershed segmentation are used to extract and calculate certain molecular characteristics from hyperspectral images. The quantification of single-cell properties is one of the key resolutions, representing the treatment efficiency in therapy of colon and rectum cancerous conditions. This was accomplished by using manually counted cells as a reference point for comparing segmentation results. The early findings acquired are conclusive for further study; thus, the extracted features will be used in the feature optimization process first, followed by neural network building of the required model.

Diversity and Chemical Library Networks of Large Data Sets

The quantification of chemical diversity has many applications in drug discovery, organic chemistry, food, and natural product chemistry, to name a few. As the size of the chemical space is expanding rapidly, it is imperative to develop efficient methods to quantify the diversity of large and ultralarge chemical libraries and visualize their mutual relationships in chemical space. Herein, we show an application of our recently introduced extended similarity indices to measure the fingerprint-based diversity of 19 chemical libraries typically used in drug discovery and natural products research with over 18 million compounds. Based on this concept, we introduce the Chemical Library Networks (CLNs) as a general and efficient framework to represent visually the chemical space of large chemical libraries providing a global perspective of the relation between the libraries. For the 19 compound libraries explored in this work, it was found that the (extended) Tanimoto index offers the best description of extended similarity in combination with RDKit fingerprints. CLNs are general and can be explored with any structure representation and similarity coefficient for large chemical libraries.

QSAR analysis of 3-pyrimidin-4-yl-oxazolidin-2-one derivatives isocitrate dehydrogenase inhibitors using Topomer CoMFA and HQSAR methods

A series of mIDH1 inhibitors derived from 3-pyrimidine-4-oxazolidin-2-ketone derivatives were studied by QSAR model to explore the key factors that inhibit mIDH1 activity. The generated model was cross-verified and non-cross-verified by Topomer CoMFA and HQSAR methods; the independent test set was verified by PLS method; the Topomer search technology was used for virtual screening and molecular design; and the Surflex-Dock method and ADMET technology were used for molecular docking, pharmacology and toxicity prediction of the designed drug molecules. The Topomer CoMFA and HQSAR cross-validation coefficients q² are 0.783 and 0.784, respectively, and the non-cross-validation coefficients r² are 0.978 and 0.934, respectively. Ten new drug molecules have been designed using Topomer search technology. The results of molecular docking and ADMET show that the newly designed drug molecules are effective. The docking situation, pharmacology and toxicity prediction results are good. The model can be used to predict the bioactivity of the same type of new compounds and their derivatives. The prediction results of molecular design, molecular docking and ADMET can provide some ideas for the design and development of novel mIDH1 inhibitor anticancer drugs, and provide certain theoretical basis of the experimental verification of new compounds in the future. Newly designed molecules after docking with corresponding proteins in the PDB library, it can explore the targets of drug molecules acting with large proteins and the related force, which is very helpful for the design of new drugs and the mechanism of drug action.

Extended similarity indices: the benefits of comparing more than two objects simultaneously. Part 1: Theory and characteristics†

Quantification of the similarity of objects is a key concept in many areas of computational science. This includes cheminformatics, where molecular similarity is usually quantified based on binary fingerprints. While there is a wide selection of available molecular representations and similarity metrics, there were no previous efforts to extend the computational framework of similarity calculations to the simultaneous comparison of more than two objects (molecules) at the same time. The present study bridges this gap, by introducing a straightforward computational framework for comparing multiple objects at the same time and providing extended formulas for as many similarity metrics as possible. In the binary case (i.e. when comparing two molecules pairwise) these are naturally reduced to their well-known formulas. We provide a detailed analysis on the effects of various parameters on the similarity values calculated by the extended formulas. The extended similarity indices are entirely general and do not depend on the fingerprints used. Two types of variance analysis (ANOVA) help to understand the main features of the indices: (i) ANOVA of mean similarity indices; (ii) ANOVA of sum of ranking differences (SRD). Practical aspects and applications of the extended similarity indices are detailed in the accompanying paper: Miranda-Quintana et al. J Cheminform. 2021. https://doi.org/10.1186/s13321-021-00504-4 . Python code for calculating the extended similarity metrics is freely available at: https://github.com/ramirandaq/MultipleComparisons .

QSAR Modeling, Molecular Docking and Molecular Dynamics Simulations Studies of Lysine-Specific Demethylase 1 (LSD1) Inhibitors as Anticancer Agents

BACKGROUND

Histone Lysine Demetylases1 (LSD1) is a promising medication to treat cancer, which plays a crucial role in epigenetic modulation of gene expression. Inhibition of LSD1with small molecules has emerged as a vital mechanism to treat cancer.

OBJECTIVE

In the present research, molecular modeling investigations, such as CoMFA, CoMFA-RF, CoMSIA and HQSAR, molecular docking and Molecular Dynamics (MD) simulations were carried out on some tranylcypromine derivatives as LSD1 inhibitors.

METHODS

The QSAR models were carried out on a series of Tranylcypromine derivatives as data set via the SYBYL-X2.1.1 program. Molecular docking and MD simulations were carried out by the MOE software and the SYBYL program, respectively. The internal and external predictability performances related to the generated models for these LSD1 inhibitors were justified by evaluating cross-validated correlation coefficient (q²), noncross- validated correlation coefficient (r²_ncv) and predicted correlation coefficient (r²_pred) of the training and test set molecules, respectively.

RESULTS

The CoMFA (q², 0.670; r²_ncv, 0.930; r²_pred, 0.968), CoMFA-RF (q², 0.694; r²_ncr, 0.926; r²_pred, 0.927), CoMSIA (q², 0.834; r²_ncv, 0.956; r²_pred, 0.958) and HQSAR models (q², 0.854; r²_ncv, 0.900; r²_pred, 0.728) for training as well as the test set of LSD1 inhibition resulted in significant findings.

CONCLUSION

These QSAR models were found to be perfect and strong with better predictability. Contour maps of all models were generated and it was proven by molecular docking studies and molecular dynamics simulation that the hydrophobic, electrostatic and hydrogen bonding fields are crucial in these models for improving the binding affinity and determining the structure-activity relationship. These theoretical results are possibly beneficial to design new strong LSD1 inhibitors with enhanced activity to treat cancer.

Sample-size dependence of validation parameters in linear regression models and in QSAR

The dependence of statistical validation parameters was investigated on the size of the sample taken in fit of multivariate linear curves. We observed that R² and related internal parameters were misleading as they overestimated the goodness-of-fit of models at small sample size. Cross-validation metrics showed correct trends. It was possible to scale the leave-one-out and the leave-many-out results close to identical by correcting the degrees of freedom of the models. y and x-randomized validation parameters were calculated and the methods provided close to identical results. We suggest to use the simplest methods in both cases. The external parameters followed correct trends with respect to the sample size, but their sensitivity differed. We plotted the Roy-Ojha metrics in 2D and we coloured them with respect to other external parameters to provide an easy classification of models. The rank correlations were calculated between the performance parameters. Up to a sample size, goodness-of-fit and robustness were distinguishable, but above a certain sample size, the parameters were redundant. The external-internal pairs were weakly correlated. Our data show that all the three aspects of validation are necessary at small sample sizes, but the internal check of robustness is not informative above a given sample size.

A computational approach for designing novel SARS-CoV-2 Mpro inhibitors: combined QSAR, molecular docking, and molecular dynamics simulation techniques

The promising compound T₂₁ for treating COVID-19 at the active site of SARS-CoV-2 M^pro.

Multicriteria decision making for evergreen problems in food science by sum of ranking differences

Finding optimal solutions usually requires multicriteria optimization. The sum of ranking differences (SRD) algorithm can efficiently solve such problems. Its principles and earlier applications will be discussed here, along with meta-analyses of papers published in various subfields of food science, such as analytics in food chemistry, food engineering, food technology, food microbiology, quality control, and sensory analysis. Carefully selected real case studies give an overview of the wide range of applications for multicriteria optimizations, using a free, easy-to-use and validated method. Results are presented and discussed in a way that helps scientists and practitioners, who are less familiar with multicriteria optimization, to integrate the method into their research projects. The utility of SRD, optionally coupled with other statistical methods such as ANOVA, is demonstrated on altogether twelve case studies, covering diverse method comparison and data evaluation scenarios from various subfields of food science.

Combined QSAR Model and Chemical Similarity Search for Novel HMG-CoA Reductase Inhibitors for Coronary Heart Disease

Background:

Coronary heart disease generally occurs due to cholesterol accumulation in the walls of the heart arteries. Statins are the most widely used drugs which work by inhibiting the active site of 3-Hydroxy-3-methylglutaryl-CoA reductase (HMGCR) enzyme that is responsible for cholesterol synthesis. A series of atorvastatin analogs with HMGCR inhibition activity have been synthesized experimentally which would be expensive and time-consuming.

Methods:

In the present study, we employed both the QSAR model and chemical similarity search for identifying novel HMGCR inhibitors for heart-related diseases. To implement this, a 2D QSAR model was developed by correlating the structural properties to their biological activity of a series of atorvastatin analogs reported as HMGCR inhibitors. Then, the chemical similarity search of atorvastatin analogs was performed by using PubChem database search.

Results and Discussion:

The three-descriptor model of charge (GATS1p), connectivity (SCH-7) and distance (VE1_D) of the molecules is obtained for HMGCR inhibition with the statistical values of R2= 0.67, RMSEtr= 0.33, R2 ext= 0.64 and CCCext= 0.76. The 109 novel compounds were obtained by chemical similarity search and the inhibition activities of the compounds were predicted using QSAR model, which were close in the range of experimentally observed threshold.

Conclusion:

The present study suggests that the QSAR model and chemical similarity search could be used in combination for identification of novel compounds with activity by in silico with less computation and effort.

Principles of QSAR Modeling

At the end of her academic career, the author summarizes the main aspects of QSAR modeling, giving comments and suggestions according to her 23 years' experience in QSAR research on environmental topics. The focus is mainly on Multiple Linear Regression, particularly Ordinary Least Squares, using a Genetic Algorithm for variable selection from various theoretical molecular descriptors, but the comments can be useful also for other QSAR methods. The need for rigorous validation, also external, and for applicability domain check to guarantee predictivity and reliability of QSAR models is particularly highlighted. The commented approach is the “predictive” one, based on chemometrics, and is usefully applied to the prioritization of environmental pollutants. All the discussed points and the author's ideas are implemented in the software QSARINS, as a legacy to the QSAR community.

Prediction the contents of fructose, glucose, sucrose, fructo-oligosaccharides and iridoid glycosides in Morinda officinalis radix using near-infrared spectroscopy

Morindae officinalis radix (MOR) is a famous Chinese herbal medicine which has long history of use in medicine and food. MOR and MOR with steaming process (PMOR) are the most commonly used forms in in clinical and health care. In order to establish a fast and mostly nondestructive quality control method for MOR, 183 beaches of MOR samples and 20 beaches of PMOR samples were collected commercially from major producing areas in Guangdong, Fujian and Guangxi Provinces of China. To predict main components of MOR, a calibration model was established based on near-infrared spectroscopy with partial least square regression. The model was optimized by compared the parameters of root mean square error of prediction (RMSEP), root mean square error of cross validation (RMSECV), coefficient of correlation (R²) and ratio of performance to deviation (RPD). Comparative studies were performed to evaluate the performance of models by different spectra preprocessing methods and different data set. The results showed that the model performance was improved with standard normal variate spectra preprocessing methods and when the data set contained both MOR and PMOR samples. A few PMOR samples were added to MOR samples data set the model predictive performance could be improved. The contents of 14 components were predicted in MOR with lower RMSEP and RMSECV, and higher R² and RPD, including fructose (12.8 mg/g, 16.3 mg/g, 0.9873, 10.10), glucose (7.28 mg/g, 8.73 mg/g, 0.9611, 6.21 sucrose (9.24 mg/g, 9.10 mg/g, 0.8419, 1.75), GF2(9.42 mg/g, 11.3 mg/g, 0.8526, 2.03), GF3(7.98 mg/g, 9.20 mg/g, 0.8756, 2.74), GF4(6.81 mg/g, 8.93 mg/g, 0.8663, 3.06), GF5(8.13 mg/g, 8.85 mg/g, 0.9001, 3.06), GF6(6.40 mg/g, 6.95 mg/g, 0.9145, 3.27), GF7(5.53 mg/g, 6.15 mg/g, 0.9195, 3.57), GF8(5.40 mg/g, 6.02 mg/g, 0.9179, 3.31), GF9(3.00 mg/g,4.35 mg/g,0.9446, 5.03),GF10(4.08 mg/g, 5.34 mg/g, 0.8983, 3.62), GF11(8.97 mg/g, 7.70 mg/g, 0.8683, 2.01) and iridoid glycosides (4.12 mg/g, 5.51 mg/g, 0.8712, 2.43). The model established in this paper could predict 14 components of MOR. The results would provide a reference method for the quality control of Chinese medical materials and their process products.

Ensemble Models Based on QuBiLS-MAS Features and Shallow Learning for the Prediction of Drug-Induced Liver Toxicity: Improving Deep Learning and Traditional Approaches

Drug-induced liver injury (DILI) is a key safety issue in the drug discovery pipeline and a regulatory concern. Thus, many in silico tools have been proposed to improve the hepatotoxicity prediction of organic-type chemicals. Here, classifiers for the prediction of DILI were developed by using QuBiLS-MAS 0-2.5D molecular descriptors and shallow machine learning techniques, on a training set composed of 1075 molecules. The best ensemble model build, E13, was obtained with good statistical parameters for the learning series, namely, the following: accuracy = 0.840, sensibility = 0.890, specificity = 0.761, Matthew's correlation coefficient = 0.660, and area under the ROC curve = 0.904. The model was also satisfactorily evaluated with Y-scrambling test, and repeated k-fold cross-validation and repeated k-holdout validation. In addition, an exhaustive external validation was also carried out by using two test sets and five external test sets, with an average accuracy value equal to 0.854 (±0.062) and a coverage equal to 98.4% according to its applicability domain. A statistical comparison of the performance of the E13 model, with regard to results and tools (e.g., Padel DDPredictor Software, Deep Learning DILIserver, and Vslead) reported in the literature, was also performed. In general, E13 presented the best global performance in all experiments. The sum of the ranking differences procedure provided a very similar grouping pattern to that of the M-ANOVA statistical analysis, where E13 was identified as the best model for DILI predictions. A noncommercial and fully cross-platform software for the DILI prediction was also developed, which is freely available at http://tomocomd.com/apps/ptoxra. This software was used for the screening of seven data sets, containing natural products, leads, toxic materials, and FDA approved drugs, to assess the usefulness of the QSAR models in the DILI labeling of organic substances; it was found that 50-92% of the evaluated molecules are positive-DILI compounds. All in all, it can be stated that the E13 model is a relevant method for the prediction of DILI risk in humans, as it shows the best results among all of the methods analyzed.

Drug design by machine-trained elastic networks: predicting Ser/Thr-protein kinase inhibitors' activities

An elastic network model (ENM) represents a molecule as a matrix of pairwise atomic interactions. Rich in coded information, ENMs are hereby proposed as a novel tool for the prediction of the activity of series of molecules, with widely different chemical structures, but a common biological activity. The new approach is developed and tested using a set of 183 inhibitors of serine/threonine-protein kinase enzyme (Plk3) which is an enzyme implicated in the regulation of cell cycle and tumorigenesis. The elastic network (EN) predictive model is found to exhibit high accuracy and speed compared to descriptor-based machine-trained modeling. EN modeling appears to be a highly promising new tool for the high demands of industrial applications such as drug and material design.

Multivariate image analysis-quantitative structure-retention relationship study of polychlorinated biphenyls using partial least squares and radial basis function neural networks

Polychlorinated biphenyls belong to a class of hazardous and environmental pollutants. Gas chromatography separation and experimental relative retention time evaluation of these compounds on a poly (94% methyl/5% phenyl) silicone-based capillary non-bonded and cross-linked column are time consuming and expensive. In this study, relative retention times were estimated using two-dimensional images of molecules based on a newly implemented rapid and simple quantitative structure retention relationship methodology. The resulting descriptors were subjected to partial least square and principal component-radial basis function neural networks as linear and nonlinear models, respectively, to attain a statistical explanation of the retention behavior of the molecules. The high numerical values of correlation coefficients and low root mean square errors in the case of the partial least square model, confirm the supremacy of this model as well as the linear dependency of images of molecules to their relative retention times. Evaluation of the best correlation model performed using internal and external tests and its good applicability domain was checked using a distance to the model in the X-Space plot. This study provides a practical and effective method for analytical chemists working with chromatographic platforms to improve predictive confidence of studies that seek to identify unknown molecules or impurities.

All-Assay-Max2 pQSAR: Activity Predictions as Accurate as Four-Concentration IC50s for 8558 Novartis Assays

Profile-quantitative structure-activity relationship (pQSAR) is a massively multitask, two-step machine learning method with unprecedented scope, accuracy, and applicability domain. In step one, a "profile" of conventional single-assay random forest regression models are trained on a very large number of biochemical and cellular pIC₅₀ assays using Morgan 2 substructural fingerprints as compound descriptors. In step two, a panel of partial least squares (PLS) models are built using the profile of pIC₅₀ predictions from those random forest regression models as compound descriptors (hence the name). Previously described for a panel of 728 biochemical and cellular kinase assays, we have now built an enormous pQSAR from 11 805 diverse Novartis (NVS) IC₅₀ and EC₅₀ assays. This large number of assays, and hence of compound descriptors for PLS, dictated reducing the profile by only including random forest regression models whose predictions correlate with the assay being modeled. The random forest regression and pQSAR models were evaluated with our "realistically novel" held-out test set, whose median average similarity to the nearest training set member across the 11 805 assays was only 0.34, comparable to the novelty of compounds actually selected from virtual screens. For the 11 805 single-assay random forest regression models, the median correlation of prediction with the experiment was only r_ext² = 0.05, virtually random, and only 8% of the models achieved our standard success threshold of r_ext² = 0.30. For pQSAR, the median correlation was r_ext² = 0.53, comparable to four-concentration experimental IC₅₀s, and 72% of the models met our r_ext² > 0.30 standard, totaling 8558 successful models. The successful models included assays from all of the 51 annotated target subclasses, as well as 4196 phenotypic assays, indicating that pQSAR can be applied to virtually any disease area. Every month, all models are updated to include new measurements, and predictions are made for 5.5 million NVS compounds, totaling 50 billion predictions. Common uses have included virtual screening, selectivity design, toxicity and promiscuity prediction, mechanism-of-action prediction, and others. Several such actual applications are described.

Effect of hydrophobic and hydrogen bonding interactions on the potency of ß-alanine analogs of G-protein coupled glucagon receptor inhibitors

G-protein coupled glucagon receptors (GCGRs) play an important role in glucose homeostasis and pathophysiology of Type-II Diabetes Mellitus (T2DM). The allosteric pocket located at the trans-membrane domain of GCGR consists of hydrophobic (TM5) and hydrophilic (TM7) units. Hydrophobic interactions with the amino acid residues present at TM5, found to facilitate the favorable orientation of antagonist at GCGR allosteric pocket. A statistically robust and highly predictive 3D-QSAR model was developed using 58 β-alanine based GCGR antagonists with significant variation in structure and potency profile. The correlation coefficient (R² ) and cross-validation coefficient (Q² ) of the developed model were found to be 0.9981 and 0.8253, respectively at the PLS factor of 8. The analysis of the favorable and unfavorable contribution of different structural features on the glucagon receptor antagonists was done by 3D-QSAR contour plots. Hydrophobic and hydrogen bonding interactions are found to be main dominating non-bonding interactions in docking studies. Presence of highest occupied molecular orbital (HOMO) in the polar part and lowest unoccupied molecular orbital (LUMO) in the hydrophobic part of antagonists leads to favorable protein-ligand interactions. Molecular mechanics/generalized born surface area (MM/GBSA) calculations showed that van der Waals and nonpolar solvation energy terms are crucial components for thermodynamically stable binding of the inhibitors. The binding free energy of highly potent compound was found to be -63.475 kcal/mol; whereas the least active compound exhibited binding energy of -41.097 kcal/mol. Further, five 100 ns molecular dynamics simulation (MD) simulations were done to confirm the stability of the inhibitor-receptor complex. Outcomes of the present study can serve as the basis for designing improved GCGR antagonists.

Multi-Level Comparison of Machine Learning Classifiers and Their Performance Metrics

Machine learning classification algorithms are widely used for the prediction and classification of the different properties of molecules such as toxicity or biological activity. the prediction of toxic vs. non-toxic molecules is important due to testing on living animals, which has ethical and cost drawbacks as well. The quality of classification models can be determined with several performance parameters. which often give conflicting results. In this study, we performed a multi-level comparison with the use of different performance metrics and machine learning classification methods. Well-established and standardized protocols for the machine learning tasks were used in each case. The comparison was applied to three datasets (acute and aquatic toxicities) and the robust, yet sensitive, sum of ranking differences (SRD) and analysis of variance (ANOVA) were applied for evaluation. The effect of dataset composition (balanced vs. imbalanced) and 2-class vs. multiclass classification scenarios was also studied. Most of the performance metrics are sensitive to dataset composition, especially in 2-class classification problems. The optimal machine learning algorithm also depends significantly on the composition of the dataset.

Intercorrelation Limits in Molecular Descriptor Preselection for QSAR/QSPR

QSAR/QSPR (quantitative structure-activity/property relationship) modeling has been a prevalent approach in various, overlapping sub-fields of computational, medicinal and environmental chemistry for decades. The generation and selection of molecular descriptors is an essential part of this process. In typical QSAR workflows, the starting pool of molecular descriptors is rationalized based on filtering out descriptors which are (i) constant throughout the whole dataset, or (ii) very strongly correlated to another descriptor. While the former is fairly straightforward, the latter involves a level of subjectivity when deciding what exactly is considered to be a strong correlation. Despite that, most QSAR modeling studies do not report on this step. In this study, we examine in detail the effect of various possible descriptor intercorrelation limits on the resulting QSAR models. Statistical comparisons are carried out based on four case studies from contemporary QSAR literature, using a combined methodology based on sum of ranking differences (SRD) and analysis of variance (ANOVA).

QSAR Analysis of Multimodal Antidepressants Vortioxetine Analogs Using Physicochemical Descriptors and MLR Modeling

BACKGROUND

Vortioxetine is a multimodal antidepressant drug with combined effects on SERT as an inhibitor, 5-HT1A as agonist and 5-HT3A as an antagonist. Series of vortioxetine analogs have been reported as multi antidepressant compounds and they block serotonin transport into the neuronal cells, activate the postsynaptic 5-HT1A receptors and eliminate the low activity of 5-HT3A receptors.

OBJECTIVE

To explore the important properties of vortioxetine analogs involved in antidepressant activity by developing 2D QSAR models.

METHODS

Selections of significant descriptors were performed by Least Absolute Shrinkage and Selection Operator (LASSO) method and, the Multiple Linear Regression (MLR) method and All Subsets and GA algorithm included in QSARINS software were used for generating QSAR models. Further, the virtual screening was performed based on bioactivity and structure similarity using the PubChem database.

RESULTS

The four descriptor model of complementary information content (CIC2), solubility (bcutp3), mass (bcutm8) and partial charge in van der Waals surface area (PEOEVSA7) of the molecules is obtained for SERT inhibition with the significant statistics of R2= 0.69, RMSEtr= 0.44, R2 ext= 0.62 and CCCext= 0.78. For 5-HT1A agonist, the two descriptor model of molecular shape (Kappm3) and van der Waals volume of the atoms (bcutv11) with R2= 0.78, RMSEtr= 0.33, R2 ext = 0.83, and CCCext= 0.87 is established. The three descriptor model of information content (IC3), solubility (bcutp9) and electronegativity (GATSe5) of the molecules with R2= 0.61, RMSEtr= 0.34, R2 ext= 0.69 and CCCext= 0.72 is obtained for 5-HT3A antagonist. The antidepressant activities of 16 virtual screened compounds were predicted using the developed models.

CONCLUSION

The developed QSAR models may be useful to predict antidepressant activity for the newly synthesized vortioxetine analogs.

On the Misleading Use of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msubsup><mml:mi>Q</mml:mi> <mml:mrow><mml:mi>F</mml:mi> <mml:mn>3</mml:mn></mml:mrow> <mml:mn>2</mml:mn></mml:msubsup> </mml:math> for QSAR Model Comparison

Quantitative Structure - Activity Relationship (QSAR) models play a central role in medicinal chemistry, toxicology and computer-assisted molecular design, as well as a support for regulatory decisions and animal testing reduction. Thus, assessing their predictive ability becomes an essential step for any prospective application. Many metrics have been proposed to estimate the model predictive ability of QSARs, which have created confusion on how models should be evaluated and properly compared. Recently, we showed that the metric <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msubsup><mml:mi>Q</mml:mi> <mml:mrow><mml:mi>F</mml:mi> <mml:mn>3</mml:mn></mml:mrow> <mml:mn>2</mml:mn></mml:msubsup> </mml:math> is particularly well-suited for comparing the external predictivity of different models developed on the same training dataset. However, when comparing models developed on different training data, this function becomes inadequate and only dispersion measures like the root-mean-square error (RMSE) should be used. The intent of this work is to provide clarity on the correct and incorrect uses of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msubsup><mml:mi>Q</mml:mi> <mml:mrow><mml:mi>F</mml:mi> <mml:mn>3</mml:mn></mml:mrow> <mml:mn>2</mml:mn></mml:msubsup> </mml:math> , discussing its behavior towards the training data distribution and illustrating some cases in which <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msubsup><mml:mi>Q</mml:mi> <mml:mrow><mml:mi>F</mml:mi> <mml:mn>3</mml:mn></mml:mrow> <mml:mn>2</mml:mn></mml:msubsup> </mml:math> estimates may be misleading. Hereby, we encourage the usage of measures of dispersions when models trained on different datasets have to be compared and evaluated.

QSAR modelling using combined simple competitive learning networks and RBF neural networks

The aim of this study was to propose a QSAR modelling approach based on the combination of simple competitive learning (SCL) networks with radial basis function (RBF) neural networks for predicting the biological activity of chemical compounds. The proposed QSAR method consisted of two phases. In the first phase, an SCL network was applied to determine the centres of an RBF neural network. In the second phase, the RBF neural network was used to predict the biological activity of various phenols and Rho kinase (ROCK) inhibitors. The predictive ability of the proposed QSAR models was evaluated and compared with other QSAR models using external validation. The results of this study showed that the proposed QSAR modelling approach leads to better performances than other models in predicting the biological activity of chemical compounds. This indicated the efficiency of simple competitive learning networks in determining the centres of RBF neural networks.

Greener chemicals for the future: QSAR modelling of the PBT index using ETA descriptors

Persistent, bioaccumulative and toxic (PBT) chemicals symbolize a group of substances that are not easily degraded; instead, they accumulate in different organisms and exhibit an acute or chronic toxicity. The limited empirical data on PBT chemicals, the high cost of testing together with the regulatory constraints and the international push for reduced animal testing motivate a greater reliance on predictive computational methods like quantitative structure-activity relationship (QSAR) models in PBT assessment. Papa and Gramatica have recently proposed a PBT index that could be computed directly from structural features. In the current study, we have modelled the experimentally derived PBT index data using an extended topological atom (ETA) along with constitutional descriptors to show the usefulness of the ETA indices in modelling the endpoint. The models developed through a double cross-validation (DCV) method gave the best results in terms of both internal and external validation metrics. The developed models were comparable in predictive quality to those previously reported. The current models were further used for consensus predictions of PBT behaviour for a set of pharmaceuticals and a set of synthetic drug-like compounds. The developed models can be used in PBT hazard screening for identification and prioritization of chemicals from the structural information alone.

Retention prediction in reversed phase high performance liquid chromatography using quantitative structure-retention relationships applied to the Hydrophobic Subtraction Model

Quantitative Structure-Retention Relationships (QSRR) methodology combined with the Hydrophobic Subtraction Model (HSM) have been utilized to accurately predict retention times for a selection of analytes on several different reversed phase liquid chromatography (RPLC) columns. This approach is designed to facilitate early prediction of co-elution of analytes, for example in pharmaceutical drug discovery applications where it is advantageous to predict whether impurities might be co-eluted with the active drug component. The QSRR model utilized VolSurf+ descriptors and a Partial Least Squares regression combined with a Genetic Algorithm (GA-PLS) to predict the solute coefficients in the HSM. It was found that only the hydrophobicity (η'H) term in the HSM was required to give the accuracy necessary to predict potential co-elution of analytes. Global QSRR models derived from all 148 compounds in the dataset were compared to QSRR models derived using a range of local modelling techniques based on clustering of compounds in the dataset by the structural similarity of compounds (as represented by the Tanimoto similarity index), physico-chemical similarity of compounds (represented by log D), the neutral, acidic, or basic nature of the compound, and the second dominant interaction between analyte and stationary phase after hydrophobicity. The global model showed reasonable prediction accuracy for retention time with errors of 30 s and less for up to 50% of modeled compounds. The local models for Tanimoto, nature of the compound and second dominant interaction approaches all exhibited prediction errors less than 30 s in retention time for nearly 70% of compounds for which models could be derived. Predicted retention times of five representative compounds on nine reversed-phase columns were compared with known experimental retention data for these columns and this comparison showed that the accuracy of the proposed modelling approach is sufficient to reliably predict the retention times of analytes based only on their chemical structures.

Is soft independent modeling of class analogies a reasonable choice for supervised pattern recognition?

A thorough survey of classification data sets and a rigorous comparison of classification methods show the unambiguous superiority of other techniques over soft independent modeling of class analogies (SIMCA – one class modeling) for classification.

Chemometric-assisted method development in hydrophilic interaction liquid chromatography: A review

With an enormous growth in the application of hydrophilic interaction liquid chromatography (HILIC), there has also been significant progress in HILIC method development. HILIC is a chromatographic method that utilises hydro-organic mobile phases with a high organic content, and a hydrophilic stationary phase. It has been applied predominantly in the determination of small polar compounds. Theoretical studies in computer-aided modelling tools, most importantly the predictive, quantitative structure retention relationship (QSRR) modelling methods, have attracted the attention of researchers and these approaches greatly assist the method development process. This review focuses on the application of computer-aided modelling tools in understanding the retention mechanism, the classification of HILIC stationary phases, prediction of retention times in HILIC systems, optimisation of chromatographic conditions, and description of the interaction effects of the chromatographic factors in HILIC separations. Additionally, what has been achieved in the potential application of QSRR methodology in combination with experimental design philosophy in the optimisation of chromatographic separation conditions in the HILIC method development process is communicated. Developing robust predictive QSRR models will undoubtedly facilitate more application of this chromatographic mode in a broader variety of research areas, significantly minimising cost and time of the experimental work.