Predictive QSAR modeling of aldose reductase inhibitors using Monte Carlo feature selection

Computational design of PARP-1 inhibitors: QSAR, molecular docking, virtual screening, ADMET, and molecular dynamics simulations for targeted drug development

Poly (ADP-ribose) polymerase-1 (PARP-1) inhibitors have shown promise in treating various cancers with homologous recombination repair deficiencies, particularly in breast and ovarian cancers harbouring BRCA1/2 mutations. This study aimed to identify and optimize novel PARP-1 inhibitors using the phthalazinone scaffold, known for forming strong and selective interactions with the active site of PARP-1. Through a combination of Quantitative Structure-Activity Relationship (QSAR) modelling, molecular docking simulations, and virtual screening, we discovered compounds with significant anticancer potential. Both the Multiple Linear Regression (MLR) and Support Vector Machines (SVM) models, utilizing four selected molecular descriptors, demonstrated high predictive efficiency for inhibitory activity (MLR: r2  = 0.944, Q2cv (cross-validated correlation coefficient) = 0.921, root mean square error (RMSE) = 0.249; SVM: r2  = 0.947, Q2cv = 0.887, RMSE = 0.245). Molecular docking studies revealed that several new compounds exhibited strong interactions with key amino acids GLY 227A, MET 229A, PHE 230A, and TYR 246A within the PARP-1 active site, similar to those observed in reference inhibitors Olaparib and AZD2461. Then, the top-ranked compound's (3a) ligand-protein complex underwent a 200 ns molecular dynamics (MD) simulation, confirming stable binding and revealing a robust set of intermolecular interactions maintained under physiological conditions.

MonteCat: A Basin-Hopping-Inspired Catalyst Descriptor Search Algorithm for Machine Learning Models

Proposing relevant catalyst descriptors that can relate the information on a catalyst's composition to its actual performance is an ongoing area in catalyst informatics, as it is a necessary step to improve our understanding on the target reactions. Herein, a small descriptor-engineered data set containing 3289 descriptor variables and the performance of 200 catalysts for the oxidative coupling of methane (OCM) is analyzed, and a descriptor search algorithm based on the workflow of the Basin-hopping optimization methodology is proposed to select the descriptors that better fit a predictive model. The algorithm, which can be considered wrapper in nature, consists of the successive generation of random-based modifications to the descriptor subset used in a regression model and adopting them depending on their effect on the model's score. The results are presented after being tested on linear and Support Vector Regression models with average cross-validation r2 scores of 0.8268 and 0.6875, respectively.

CADD Studies in the Discovery of Potential ARI (Aldose Reductase Inhibitors) Agents for the Treatment of Diabetic Complications

The lack of currently available drugs for treating diabetes complications has stimulated our interest in finding new Aldose Reductase inhibitors (ARIs) with more beneficial biological properties. One metabolic method uses aldose reductase inhibitors in the first step of the polyol pathway to control excess glucose flux in diabetic tissues. Computer-aided drug discovery (CADD) is key in finding and optimizing potential lead substances. AR inhibitors (ARI) have been widely discussed in the literature. For example, Epalrestat is currently the only ARI used to treat patients with diabetic neuropathy in Japan, India, and China. Inhibiting R in patients with severe to moderate diabetic autonomic neuropathy benefits heart rate variability. AT-001, an AR inhibitor, is now being tested in COVID-19 to see how safe and effective it reduces inflammation and cardiac damage. In summary, these results from animal and human studies strongly indicate that AR can cause cardiovascular complications in diabetes. The current multi-center, large-scale randomized human study of the newly developed powerful ARI may prove its role in diabetic cardiovascular disease to establish therapeutic potential. During the recent coronavirus disease (COVID-19) outbreak in 2019, diabetes and cardiovascular disease were risk factors for severely negative clinical outcomes in patients with COVID19. New data shows that diabetes and obesity are among the strongest predictors of COVID-19 hospitalization. Patients and risk factors for severe morbidity and mortality of COVID- 19.

An improved opposition-based crow search algorithm for biodegradable material classification

The development of a reliable quantitative structure-activity relationship (QSAR) classification model with a small number of molecular descriptors is a crucial step in chemometrics. In this study, an improvement of crow search algorithm (CSA) is proposed by adapting the opposite-based learning (OBL) approach, which is named as OBL-CSA, to improve the exploration and exploitation capability of the CSA in quantitative structure-biodegradation relationship (QSBR) modelling of classifying the biodegradable materials. The results reveal that the performance of OBL-CSA not only manifest in improving the classification performance, but also in reduced computational time required to complete the process when compared to the standard CSA and other four optimization algorithms tested, which are the particle swarm algorithm (PSO), black hole algorithm (BHA), grey wolf algorithm (GWA), and whale optimization algorithm (WOA). In conclusion, the OBL-CSA could be a valuable resource in the classification of biodegradable materials.

Using machine learning techniques to develop prediction models for detecting unpaid credit card customers

Customer behavior prediction is gaining more importance in the banking sector like in any other sector recently. This study aims to propose a model to predict whether credit card users will pay their debts or not. Using the proposed model, potential unpaid risks can be predicted and necessary actions can be taken in time. For the prediction of customers’ payment status of next months, we use Artificial Neural Network (ANN), Support Vector Machine (SVM), Classification and Regression Tree (CART) and C4.5, which are widely used artificial intelligence and decision tree algorithms. Our dataset includes 10713 customer’s records obtained from a well-known bank in Taiwan. These records consist of customer information such as the amount of credit, gender, education level, marital status, age, past payment records, invoice amount and amount of credit card payments. We apply cross validation and hold-out methods to divide our dataset into two parts as training and test sets. Then we evaluate the algorithms with the proposed performance metrics. We also optimize the parameters of the algorithms to improve the performance of prediction. The results show that the model built with the CART algorithm, one of the decision tree algorithm, provides high accuracy (about 86%) to predict the customers’ payment status for next month. When the algorithm parameters are optimized, classification accuracy and performance are increased.

QSAR model for predicting neuraminidase inhibitors of influenza A viruses (H1N1) based on adaptive grasshopper optimization algorithm

High-dimensionality is one of the major problems which affect the quality of the quantitative structure-activity relationship (QSAR) modelling. Obtaining a reliable QSAR model with few descriptors is an essential procedure in chemometrics. The binary grasshopper optimization algorithm (BGOA) is a new meta-heuristic optimization algorithm, which has been used successfully to perform feature selection. In this paper, four new transfer functions were adapted to improve the exploration and exploitation capability of the BGOA in QSAR modelling of influenza A viruses (H1N1). The QSAR model with these new quadratic transfer functions was internally and externally validated based on MSE_train, Y-randomization test, MSE_test, and the applicability domain (AD). The validation results indicate that the model is robust and not due to chance correlation. In addition, the results indicate that the descriptor selection and prediction performance of the QSAR model for training dataset outperform the other S-shaped and V-shaped transfer functions. QSAR model using quadratic transfer function shows the lowest MSE_train. For the test dataset, proposed QSAR model shows lower value of MSE_test compared with the other methods, indicating its higher predictive ability. In conclusion, the results reveal that the proposed QSAR model is an efficient approach for modelling high-dimensional QSAR models and it is useful for the estimation of IC₅₀ values of neuraminidase inhibitors that have not been experimentally tested.

Design of potential anti-tumor PARP-1 inhibitors by QSAR and molecular modeling studies

Poly ADP-ribose polymerase-1 (PARP-1) inhibitors have been recognized as new agents for the treatment of patients with breast cancer type 1 (BRCA1) disorders. The quantitative structure-activity relationships (QSAR) technique was used in order to achieve the required medicines for anticancer activity easier and faster. In this study, the QSAR method was developed to predict the half-maximal inhibitory concentration (IC₅₀) of 51 1H-benzo[d]immidazole-4-carboxamide derivatives by genetic algorithm-multiple linear regression (GA-MLR) and least squares-support vector machine (LS-SVM) methods. Results in the best QSAR model represented the coefficient of leave-one-out cross-validation (Q _cv ² ) = 0.971, correlation coefficient (R²) = 0.977, Fisher parameter (F) = 259.016 and root mean square error (RMSE) = 0.095, respectively, which indicated that the LS-SVM model had a good potential to predict the pIC₅₀ (9 - log(IC₅₀ nM)) values compared with other modeling methods. Also, molecular docking evaluated interactions between ligands and enzyme and their free energy of binding were calculated and used as descriptors. Molecular docking and the QSAR study completed each other. The results represented that the final model can be useful to design some new inhibitors. So, the knowledge of the QSAR modeling and molecular docking was used in pIC₅₀ prediction and 51 new compounds were developed as PARP-1 inhibitors that 9 compounds had the best-proposed values for pIC₅₀. The maximum enhancement of the inhibitory activity of compounds was 33.394%.

Small-Angle X-ray Scattering Study of Protein Complexes with Tea Polyphenols

Exploration of the structure of protein complexes, especially the change in conformation and aggregation behavior of proteins upon ligand binding, is crucial to clarify their bioactivities at the molecular level. We applied solution small-angle X-ray scattering (SAXS) to study the complex structure of bovine serum albumin (BSA) and trypsin binding with tea polyphenols, that is, catechin and epigallocatechin gallate (EGCG). We found that tea polyphenols can steadily promote the aggregation of proteins and protein complexes through their bridging effect. The numbers of proteins in the complexes and in the aggregates of complexes are extracted from SAXS intensity profiles, and their dependences as a function of the molar ratio of polyphenol to protein are discussed. EGCG has stronger capability than catechin to promote complex formation and further aggregation, and the aggregates of complexes have a denser core with a relatively smooth surface. The aggregates induced by catechin are loosely packed with a rough surface. BSA shows higher stability than trypsin in the formation of complex with a well-folded conformation. The synergistic unfolding of trypsin results in larger aggregates in the mixtures with more tea polyphenols. The binding affinity and number of tea polyphenols bound to each protein are further determined using fluorescence spectroscopy. The structure of protein complexes explored in this work is referable in the preparation of protein complex-based particles and the understanding of polyphenol-induced formation and further aggregation of protein complexes.

A new adaptive L1-norm for optimal descriptor selection of high-dimensional QSAR classification model for anti-hepatitis C virus activity of thiourea derivatives

A high-dimensional quantitative structure-activity relationship (QSAR) classification model typically contains a large number of irrelevant and redundant descriptors. In this paper, a new design of descriptor selection for the QSAR classification model estimation method is proposed by adding a new weight inside L1-norm. The experimental results of classifying the anti-hepatitis C virus activity of thiourea derivatives demonstrate that the proposed descriptor selection method in the QSAR classification model performs effectively and competitively compared with other existing penalized methods in terms of classification performance on both the training and the testing datasets. Moreover, it is noteworthy that the results obtained in terms of stability test and applicability domain provide a robust QSAR classification model. It is evident from the results that the developed QSAR classification model could conceivably be employed for further high-dimensional QSAR classification studies.

Joint L2,1 Norm and Fisher Discrimination Constrained Feature Selection for Rational Synthesis of Microporous Aluminophosphates

Feature selection has been regarded as an effective tool to help researchers understand the generating process of data. For mining the synthesis mechanism of microporous AlPOs, this paper proposes a novel feature selection method by joint l_2,1 norm and Fisher discrimination constraints (JNFDC). In order to obtain more effective feature subset, the proposed method can be achieved in two steps. The first step is to rank the features according to sparse and discriminative constraints. The second step is to establish predictive model with the ranked features, and select the most significant features in the light of the contribution of improving the predictive accuracy. To the best of our knowledge, JNFDC is the first work which employs the sparse representation theory to explore the synthesis mechanism of six kinds of pore rings. Numerical simulations demonstrate that our proposed method can select significant features affecting the specified structural property and improve the predictive accuracy. Moreover, comparison results show that JNFDC can obtain better predictive performances than some other state-of-the-art feature selection methods.

Construction of explicit models to correlate the structure and the inhibitory activity of aldose reductase: Flavonoids and sulfonyl-pyridazinones as inhibitors

The correlation between binding energies and bioactivities is the core of structure-based computer-aided drug design. However, many models to address this correlation are still strongly system-dependent at current stage. We constructed two explicit models to correlate the binding energies with the inhibitory activities of flavonoids and sulfonyl-pyridazinones as inhibitors of aldose reductase. The introduction of multiple complex states comprised of protein, coenzyme, substrate, and inhibitor can remarkably improve the correlation coefficients, compared with that using single complex state. Recombination of energy terms from complex structures and molecular descriptors of inhibitors can further improve the correlation. The explicit models provide correlation coefficients of 0.90 and 0.92 for flavonoids and sulfonyl-pyridazinones, respectively. These models also steadily present the contribution from each energy term and the favorite of protein-inhibitor complex states. Meanwhile, we also observed that some inhibitors can accommodate alternative sites out of the conserved binding pocket at the presence/absence of coenzyme and substrate. It is responsible for the remarkable change in the binding energies and thus significantly influences the correlation between the structures and the inhibitory activities. Overall, this work presents a rational way to construct reliable explicit models through the combination of multiple physically accessible complex states, even though each of them only bears marginal information related to their activities.

High-dimensional QSAR modelling using penalized linear regression model with L1/2-norm

In high-dimensional quantitative structure-activity relationship (QSAR) modelling, penalization methods have been a popular choice to simultaneously address molecular descriptor selection and QSAR model estimation. In this study, a penalized linear regression model with L1/2-norm is proposed. Furthermore, the local linear approximation algorithm is utilized to avoid the non-convexity of the proposed method. The potential applicability of the proposed method is tested on several benchmark data sets. Compared with other commonly used penalized methods, the proposed method can not only obtain the best predictive ability, but also provide an easily interpretable QSAR model. In addition, it is noteworthy that the results obtained in terms of applicability domain and Y-randomization test provide an efficient and a robust QSAR model. It is evident from the results that the proposed method may possibly be a promising penalized method in the field of computational chemistry research, especially when the number of molecular descriptors exceeds the number of compounds.

Discovery of novel 1,2,3-triazole derivatives as anticancer agents using QSAR and in silico structural modification

Considerable attention has been given on the search for novel anticancer drugs with respect to the disease sequelae on human health and well-being. Triazole is considered to be an attractive scaffold possessing diverse biological activities. Structural modification on the privileged structures is noted as an effective strategy towards successful design and development of novel drugs. The quantitative structure–activity relationships (QSAR) is well-known as a powerful computational tool to facilitate the discovery of potential compounds. In this study, a series of thirty-two 1,2,3-triazole derivatives (1–32) together with their experimentally measured cytotoxic activities against four cancer cell lines i.e., HuCCA-1, HepG2, A549 and MOLT-3 were used for QSAR analysis. Four QSAR models were successfully constructed with acceptable predictive performance affording R_CV ranging from 0.5958 to 0.8957 and RMSE_CV ranging from 0.2070 to 0.4526. An additional set of 64 structurally modified triazole compounds (1A–1R, 2A–2R, 7A–7R and 8A–8R) were constructed in silico and their predicted cytotoxic activities were obtained using the constructed QSAR models. The study suggested crucial moieties and certain properties essential for potent anticancer activity and highlighted a series of promising compounds (21, 28, 32, 1P, 8G, 8N and 8Q) for further development as novel triazole-based anticancer agents.