Prediction of concentration immediately dangerous to life or health of benzene and its derivatives based on quantitative structure-activity relationship / 环境与职业医学

Background With the increasing exposure to hazardous chemicals in the workplace and frequency of occupational injuries and occupational safety accidents, the acquisition of occupational exposure limits of hazardous chemicals is imminent. Objective To obtain more unknown immediately dangerous to life or health (IDLH) concentrations of hazardous chemicals in the workplace by exploring the application of quantitative structure-activity relationship (QSAR) prediction method to IDLH concentrations, and to provide a theoretical basis and technical support for the assessment and prevention of occupational injuries. Methods QSAR was used to correlate the IDLH values of 50 benzene and its derivatives with the molecular structures of target compounds. Firstly, affinity propagation algorithm was applied to cluster sample sets. Secondly, Dragon 2.1 software was used to calculate and pre-screen 537 molecular descriptors. Thirdly, the genetic algorithm was used to select six characteristic molecular descriptors as dependent variables and to construct a multiple linear regression model (MLR) and two nonlinear models using support vector machine (SVM) and artificial neural network (ANN) respectively. Finally, model performance was evaluated by internal and external validation and Williams diagram was drawn to determine the scopes of selected models. Results The ANN model results showed that \begin{document}$ {R}_{\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}}

Applications of QSAR in Toxicological Risk Assessment of Medical Devices / 中国医疗器械杂志

The chemical characterization analysis of a medical device often results in chemical substances with unknown toxicities. While identification of each individual toxicity could result in a time-consuming hurdle with tremendous labor and financial burden, quantitative structure-activity relationship (QSAR) is of great significance for toxicity risk assessment of such chemical substances. By establishing quantitative relationship between the molecular structures or active groups of similar chemical compounds with their biological activities, QSAR can be utilized to predict the toxicity of such target compounds with significantly reduced cost and time. In this article, the authors generally summarized the mechanisms of QSAR approaches, current applications of QSAR modeling in the field of medical device, an introduction of the characteristics of publicly and commercially-available QSAR software, and briefly explored future trends of QSAR modeling in medical device toxicological risk assessment. The utilization of QSAR would undoubtedly further advance the toxicological risk assessment of medical devices.

Application value of magnetic resonance T2mapping in the diagnosis of knee osteoarthritis / 中国基层医药

Objective:To investigate the application value of magnetic resonance T2mapping in the diagnosis of knee osteoarthritis (KOA).Methods:The MRI data of the knees of 148 patients with KOA who underwent diagnosis and treatment between January 2017 and December 2020 in Benxi Central Hospital (KOA group) and 30 healthy volunteers (control group) were retrospectively analyzed. T2 values of cartilage in each sub-region of the knee were measured, grouped, and statistically analyzed.Results:There was no significant difference in the T2 value of cartilage in each sub-region of the knee between male and female patients in mild and severe KOA groups (all P > 0.05). T2 values in the medial anterior, middle, and posterior areas of the tibia, lateral anterior, middle and posterior areas of the tibia, medial middle, posterior and lateral areas of the femur, and lateral posterior area of the femur were (44.47 ± 2.35) ms, (46.52 ± 3.12) ms, (45.47 ± 2.40) ms, (43.68 ± 2.12) ms, (46.33 ± 3.36) ms, (43.92 ± 3.42) ms, (43.58 ± 2.40) ms, (45.53 ± 3.91) ms, (44.36 ± 3.15) ms, (46.41 ± 3.04) ms, respectively in the control group. They were (49.56 ± 2.05) ms, (51.67 ± 2.38) ms, (50.47 ± 2.53) ms, (48.68 ± 3.05) ms, (51.33 ± 4.62) ms, (48.92 ± 2.53) ms, (48.58 ± 3.15) ms, (50.53 ± 3.72) ms, (48.36 ± 2.41) ms, and (51.41 ± 3.64) ms, respectively in the mild KOA group, and (53.47 ± 2.46) ms, (56.52 ± 3.57) ms, (54.85 ± 2.89) ms, (52.68 ± 3.57) ms, (56.33 ± 3.91) ms, (52.92 ± 3.04) ms, (53.58 ± 3.36) ms, (55.53 ± 3.42) ms, (52.36 ± 4.13) ms, and (56.41 ± 3.56) ms, respectively in the severe KOA group. There were significant differences in abovementioned indices among the three groups ( F = 38.768, 39.412, 38.981, 40.432, 38.416, 38.635, 38.347, 40.712, 38.158, 39.418, all P < 0.05). Conclusion:The T2 value of knee cartilage in patients with KOA is unrelated to gender and related to the severity of the disease. Magnetic resonance T2 mapping can help diagnose KOA, and provide information about the changes in cartilage components of patients with early KOA.

Identification of small molecule sirtuin-1 inhibitors for treating acute myeloid leukemia based on molecular docking, quantitative structure-activity relationships and molecular dynamics / 药学学报

The purpose of this study was to discover novel inhibitors of sirtuin-1 (SIRT1) that could be used in the treatment of acute myeloid leukemia (AML). Eight potential SIRT1 inhibitors were identified from 231 511 natural drug-like molecules by virtual screening-based molecular docking and molecular mechanics-generalized Born surface area (MM-GBSA) calculation of binding free energies. Using existing SIRT1 inhibitor molecules as training and test sets, a series of quantitative structure-activity relationship models were established, and the best quantitative structure-activity relationship (QSAR) model was used to predict the IC50 of these 8 potential inhibitor molecules for SIRT1. Subsequently, molecular dynamics simulations were performed to verify the binding mode and stability of these complexes of potential inhibitors and SIRT1 protein. Finally, the activity of these potential SIRT1 inhibitors was verified by cell proliferation assays of OCI-AML2, OCI-AML3 and MV4-11 cells and SIRT1 enzyme activity assays, and it was found that 5 compounds could inhibit AML cell proliferation. Among them, the most active compound, ZINC000001774455, had an IC50 of 2.29 ± 0.09 μmol·L-1 with OCI-AML2 cells, and at a concentration of 1 μmol·L-1, the inhibitory ratio of this compound on SIRT1 protein activity was 65.33%. ZINC000001774455 can be used as a lead compound for the development of new AML treatments.

Evaluation genotoxicity of acetylnerolin based on quantitative structure-activity relationship model and in vitro tests / 中国药理学与毒理学杂志

OBJECTIVE To evaluate the genotoxicity of naproxen (NPX) impurities acetylnerolin (Ace). METHODS The genotoxicity of Ace was predicted by ADMET, Derek and Sarah with the quanti?tative structure-activity relationship (QSAR). The chromosomal aberration and bacterial reverse-muta?tion (Ames) tests were performed to verify the above results. In chromosomal aberration tests, CHL cells were incubated with Ace 10, 20 and 40 mg · L-1 for 4 h in the presence or absence of metabolic activation system solution (S9 mix). Methyl methane sulfonate (MMS) 20 mL · L-1 without S9 mix and cyclophosphamide (CP) 12 mg · L-1 with S9 mix served as positive control. The number of chromo?somes in each aberrant metaphase (including fissure, exchange, ring, break and polyploid) was counted and recorded, when the distortion rate less than 5％was considered negative and more than 10％was considered positive. In Ames test, the potential mutagenicity was evaluated using five strains of S. typhimurium ( TA97,TA98,TA100,TA102 and TA1535). They were treated with Ace 5, 25, 125 and 625μg per plate with or without S9 mix and incubated for 48-72 h. When without S9 mix, Dexon 50μg per plate served as positive control for TA97 and TA98, MMS 2.0μL per plate served as positive control for TA100 and TA102, and sodium azide 1.5μg per plate served as positive control for TA1535. When with S9 mix, 2-AF 100 μg per plate served as positive control for TA97, TA98 and TA100, 1, 8-dihydroxyanthraquinone (100μg per plate) served as positive control for TA102 and CP 50μg per plate served as positive control for TA1525. When the number of colonies was at least two-fold that of the negative control, the compound was considered mutagenic. RESULTS Although the Derek and Sarah software predicted that the NPX impurities were not genotoxic, ADMET data showed that Ace could induce chromosomal aberrations. The distortion rate of Ace 40 mg · L-1 was greater than 5％, but less than 10％. The distortion rate of Ace was less than 5％when<20 mg·L-1. Consistent with the results of ADMET, Ace might induce chromosomal aberrations. Ames test results showed that Ace did not signifi?cantly increase the number of bacteria (5-625μg per plate) compared with the negative control. Contrary to the ADMET results, Ace had no mutagenicity. CONCLUSION Ace has potential chromosomal muta?genicity. For life-long usage of NPX, the content of Ace should be reduced from 0.15％of conventional impurities to 0.015％.

Synthesis, activity determination of tropane compounds as muscarinic M3 receptor antagonists and quantitative relationship between their molecular structure and activity / 上海交通大学学报（医学版）

Objective • To provide theoretical guidance for the design of molecules with high activity by building the quantitative structure-activity relationship (QSAR) model for tropane compounds as muscarinic M3 receptor antagonists. Methods • Six compounds (J1-J6) were prepared with 3α-hydroxy-tropane (J0) as the starting material by modifying the structure in C-3α position of the tropane skeleton. The antagonistic activity of new tropane compounds to muscarinic M3 receptors on tracheal rings of guinea pigs was evaluated by functional assays in vitro. The antagonistic parameters (pA2) of new tropane compounds prepared in this paper and former studies were applied to construct the QSAR model. The information about structural optimization was acquired by analyzing the effect of steric field and electrostatic field of model on activity of tropane compounds. Results • The six new tropane compounds showed obvious antagonistic activity against M3 receptors. Among them, J4 had the greatest activity (pA2=7.992). The cross-validation correlation coefficient squared (q2) and the non-cross-validated correlation coefficient squared (r2) of the QSAR model were 0.585 and 0.993, respectively. Conclusion • The antagonistic activity to muscarinic M3 receptors can be obviously improved, when the conjugating extent of π-bonds is large and O or N atoms participate in conjugation on the rings in the R-substituting group at C-3α position of the compounds.

Synthesis, activity determination of tropane compounds as muscarinic M3 receptor antagonists and quantitative relationship between their molecular structure and activity / 上海交通大学学报（医学版）

Objective · To provide theoretical guidance for the design of molecules with high activity by building the quantitative structure-activity relationship (QSAR) model for tropane compounds as muscarinic M3 receptor antagonists. Methods · Six compounds (J1-J6) were prepared with3α-hydroxy-tropane (J0) as the starting material by modifying the structure in C-3α position of the tropane skeleton. The antagonistic activity of new tropane compounds to muscarinic M3 receptors on tracheal rings of guinea pigs was evaluated by functional assays in vitro. The antagonistic parameters (pA2) of new tropane compounds prepared in this paper and former studies were applied to construct the QSAR model. The information about structural optimization was acquired by analyzing the effect of steric field and electrostatic field of model on activity of tropane compounds. Results · The six new tropane compounds showed obvious antagonistic activity against M3 receptors. Among them, J4 had the greatest activity (pA2=7.992). The cross-validation correlation coefficient squared (q2) and the non-cross-validated correlation coefficient squared (r2) of the QSAR model were 0.585 and 0.993, respectively.Conclusion · The antagonistic activity to muscarinic M3 receptors can be obviously improved, when the conjugating extent of π-bonds is large and O or N atoms participate in conjugation on the rings in the R-substituting group at C-3α position of the compounds.

Dipole Moment, Solvation Energy, and Ovality Account for the Variations in the Biological Activity of HIV-1 Reverse Transcriptase Inhibitor Fragments

Objective: A computational approach was employed to determine the interaction of molecular descriptors and the biological activity of the different fragments of HIV-1 reverse transcriptase inhibitors (RTIs). Methods: Using multiple linear regression analysis and leave-one-out validation method, a quantitative structure activity relationship (QSAR) model was developed to relate the biological activity (log IC50) of the different fragment-sized compounds against HIV-1 RT(WT) DNA-dependent DNA polymerase and molecular descriptors of these compounds. Results: QSAR model identified dipole moment, solvation energy, and ovality of fragment-sized compounds to confer reverse transcriptase inhibitory action. A highly significant correlation with log P, molecular weight, polarizability, molecular energy, zero-point energy, constant volume heat capacity at 298 K, and entropy was identified to account for the variations in the potency of RTIs. An increase in ovality, log P, and molecular weight of the fragment-sized compound renders a more active reverse transcriptase inhibition. Conclusion: The quality of the established QSAR model has been validated and demonstrates its potential as a tool for computational design and synthesis of next generation RTIs.

Natural radioprotectors and their impact on cancer drug discovery

Cancer is a complex multifaceted illness that affects different patients in discrete ways. For a number of cancers the use of chemotherapy has become standard practice. Chemotherapy is a use of cytostatic drugs to cure cancer. Cytostatic agents not only affect cancer cells but also affect the growth of normal cells; leading to side effects. Because of this, radiotherapy gained importance in treating cancer. Slaughtering of cancerous cells by radiotherapy depends on the radiosensitivity of the tumor cells. Efforts to improve the therapeutic ratio have resulted in the development of compounds that increase the radiosensitivity of tumor cells or protect the normal cells from the effects of radiation. Amifostine is the only chemical radioprotector approved by the US Food and Drug Administration (FDA), but due to its side effect and toxicity, use of this compound was also failed. Hence the use of herbal radioprotectors bearing pharmacological properties is concentrated due to their low toxicity and efficacy. Notably, in silico methods can expedite drug discovery process, to lessen the compounds with unfavorable pharmacological properties at an early stage of drug development. Hence a detailed perspective of these properties, in accordance with their prediction and measurement, are pivotal for a successful identification of radioprotectors by drug discovery process.

3D-QSAR of Thiazide 11β-Hydroxyl Steroid Dehydrogenase Inhibitors / 中国药师

Objective:To establish a three-dimensional quantitative structure-activity relationship (3D-QSAR) model for thiazide 11β-hydroxy steroid dehydrogenase ( HSD) inhibitors in order to perform structure modification and find thiazide 11β-HSD inhibitors with more activity. Methods: The 3D-QSAR model of thiazine derivatives was constructed by the method of comparative molecular force field analysis, and the model was validated by using a molecular docking method. Results:An accurate 3D-QSAR model of 11β-HSD inhibitors was obtained (CoMFA:q2 =0. 346, r2 =0. 850, where q2 was the cross-validation coefficient and r2 was the non-cross validation coefficient) . Conclusion:The results provide important theoretical basis for the rational design of novel thiazide 11β-HSD inhibitors.

Quantitative structure-activity relationship prediction of carcinogenicity of N-nitroso compounds based on category approach and read-across / 中华预防医学杂志

Objective@#New quantitative structure-activity relationship (QSAR) method was used to predict N-nitroso compounds (NOCs) carcinogenicity. This could provide evidences for health risk assessment of the chemicals.@*Methods@#Total 74 chemical substances of NOCs were included as target chemicals for this validation study by using QSAR Toolbox based on category approach and read-across. The included 74 NOCs were categorized and subcategorized respectively using "Organic functional groups, Norbert Haider " profiler and "DNA binding by OASIS V.1.1" profiler. Carcinogenicity of rat were used as target of prediction, the carcinogenicity@*results@#of analogues in chemical categories were cross-read to obtain the carcinogenic predictive results of the target chemicals. Results 74 NOCs included 26 nonclic N-nitrosamines, 24 cyclic N-nitrosamines and 24 N-nitrosamides The sensitivity, specificity and concordance of the category approach and read-across for predicting carcinogenicity of 74 NOCs were 75% (48/64), 70%(7/10) and 74% (55/74) respectively. The concordance for noncyclic N-nitrosamines, cyclic N-nitrosamines and N-nitrosamides were 88% (23/26), 71% (17/24) and 63% (15/24) respectively.@*Conclusion@#QSAR based on category approach and read-across is good for prediction of NOCs carcinogenicity, and can be used for high-throughput qualitative prediction of NOCs carcinogenicity.

In silico investigation of agonist activity of a structurally diverse set of drugs to hPXR using HM-BSM and HM-PNN / 华中科技大学学报（医学）（英德文版）

The human pregnane X receptor (hPXR) plays a critical role in the metabolism, transport and clearance of xenobiotics in the liver and intestine. The hPXR can be activated by a structurally diverse of drugs to initiate clinically relevant drug-drug interactions. In this article, in silico investigation was performed on a structurally diverse set of drugs to identify critical structural features greatly related to their agonist activity towards hPXR. Heuristic method (HM)-Best Subset Modeling (BSM) and HM-Polynomial Neural Networks (PNN) were utilized to develop the linear and non-linear quantitative structure-activity relationship models. The applicability domain (AD) of the models was assessed by Williams plot. Statistically reliable models with good predictive power and explain were achieved (for HM-BSM, r (2)=0.881, q LOO (2) =0.797, q EXT (2) =0.674; for HM-PNN, r (2)=0.882, q LOO (2) =0.856, q EXT (2) =0.655). The developed models indicated that molecular aromatic and electric property, molecular weight and complexity may govern agonist activity of a structurally diverse set of drugs to hPXR.

QSAR Study on Toxicity of Chemical Components of Chinese Materia Medica and Acute Toxicity of Rats / 中国中医药信息杂志

Objective To study computer toxicity prediction technology and predict the acute toxicity of Chinese materia medica; To provide a new way and method for safety evaluation of traditional Chinese medicine. Methods First, Mold2 software (version 2.0.0) was used to calculate molecular descriptors of 7409 chemical components. After preliminary screening of molecular descriptors, quantitative structure-activity relationship (QSAR) models were built up with Random Forest (RF) for screening the optimum prediction model. From the 83 kinds of toxic Chinese materia medica in Chinese Pharmacopoeia (2010 edition), acute toxicity of 60 kinds of Chinese materia medica reported from monomer structure (1692 chemical components) were under prediction.Results Totally 7409 pieces of data were obtained. When the descriptors were 52, RF modeling accuracy and Kappa were the highest, 0.712 and 0.436 respectively. Compound clusters were divided into 3 types according to optimum molecule descriptors (52). The accuracy and Kappa of the optimum model for the first type of compounds were 0.666 and 0.476 respectively; the accuracy and Kappa of the optimum model for the second type of compounds were 0.804 and 0.381 respectively; the accuracy and Kappa of the optimum model for the third type of compounds were 0.709 and 0.373 respectively. It was predicted that 60 kinds of Chinese materia medica containing 0 violent toxic compound, 2 high toxic compounds, 172 medium toxic compounds and 1518 low toxic compound.Conclusion QSAR model for prediction study on acute toxicity of chemical components of Chinese mareria medica can provide references combination medication and experimental studies.

Quantitative structure-activity relationship of synthetic pyrethroids and antibody activity / 中国免疫学杂志

Objective:To research on the immune recognition mechanism of synthetic pyrethroids and generic specific antibody.Methods:We studied on quantitative structure-activity relationship ( QSAR ) of synthetic pyrethroids and their analogs as well as antibody activity ( IC50:fifty percent inhibition concentration ) using stepwise multiple linear regression method.Based on calculating structure descriptors of synthetic pyrethroids and their analogs , two-demensional QSAR ( 2D-QSAR ) model was established.The main factors affecting antibody activity were screened using 2D-QSAR,and predictive ability of QSAR models were evaluated by the method of leave-one-out( LOO) cross-validation.Meanwhile, the structure parameters of synthetic pyrethroid fragments were calculated and then analyzed using partial least squares ( PLS) assay.And then hologram QSAR ( H-QSAR) model was constructed on molecular substructure and antibody activity.The fragments contribution to antibody activity were illustrated by encoding different colors.Results:Decision coefficent (R2) of 2D-QSAR model and HQSAR model were 0.920 and 0.917 individually,cross-validation coefficient ( Q2 ) of two QSAR models were 0.875 and 0.660 respectively ,which showed two models had good predictive abil-ity.The result from 2D-QSAR model was also obtained that smaller was hydrophobicity of pyrethroids , easier was recognized by antibody.In addition,the optimum HQSAR model was constructed after we tried many combinations of these parameters .The fragment size in optimum HQSAR model was between 4 to 10,a hologram length was 61,optimum principle component was 4,and the fragment type of B/C/Ch was selected.However ,the fingerprint encoded results of synthetic pyrethroids weren′t consistent completely with exper-imental IC50 values.Conclusion:Hydrophobicity of synthetic pyrethroids is the largest correlation factors in antibody recognization .

A comparative 2D QSAR study on a series of hydroxamic acid-based histone deacetylase inhibitors vis-à-vis comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA).

A quantitative structure-activity relationship (QSAR) study was performed on a series of indole amide analogues reported by Dai et al. [Bioorg Med Chem Lett (2003), 13, 1897-1901] to act as histone deacetylase (HDAC) inhibitors. The multiple regression analysis (MRA) revealed a model showing the significant dependence of the activity on molar refractivity (MR) and global topological charge index (GTCI) of the compounds, suggesting that inhibition of the HDAC by this series of compounds might involve the dispersion interaction with the receptor, where charge transfer between pairs of atoms might greatly help to polarize the molecule. The MRA results were then compared with those obtained by Guo et al. [Bioorg Med Chem (2005), 13, 5424-5434] by comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). It was found that MRA gave as good results and had as good predictive ability as CoMFA and CoMSIA. Besides, MRA was also able to throw the light on the physicochemical properties of the molecules that were involved in drug-receptor interactions, while CoMFA and CoMSIA could not. The dispersion interaction between the molecule and the active site of the receptor is suggested to be the main interaction.

Novel Insight into the Structural Requirements of P70S6K Inhibition Using Group-based Quantitative Structure Activity Relationship (GQSAR).

Dysregulation of P70 ribosomal S6 kinase (P70S6K) has been observed in many cancers; therefore, the design of new molecules targeting p70S6K of paramount importance in cancer therapy. The current study employed a group-based quantitative structure-activity relationship (GQSAR) to develop global QSAR models capable of predicting the bioactivity of P70S6K inhibitors. A wide variety of chemical structures and biological activities (half maximal inhibitory concentration) of P70S6K inhibitors were collected from the binding database website. Compounds were classified into various chemical groups and then fragmented into R1, R2, and R3 fragments based on certain pharmacophoric features required for ligand-target biointeractions. Different two-dimensional fragment-based descriptors were calculated for each fragment. The dataset was then divided into a training set (n=40) and a test set (n=10) using a sphere exclusion algorithm. Multiple linear regressions coupled with simulated annealing or stepwise regression resulted in model A (r2=0.92) and model B (r2=0.87), respectively. Leave-one-out validation showed that models A and B have internal predictive abilities of 72% and 61%, respectively. External validation indicated that both models are robust, with squared cross-correlation coefficients of the training set (pred-r2) of 0.87 and 0.89, respectively. The developed GQSAR models indicate that fragment R3 plays a key role in activity variation (65%) with sound contribution of five-membered rings (5 chain count), aromatic carbons (SaaaCE-index), and aromatic nitrogens (SaaNcount). In contrast, fragments R1 and R2 together contribute 35% of activity variation, suggesting that sulfur atoms (Sulfur count) and hydrophobic threemembered rings (chi3 chain) at R1 are preferable for inhibitory activity.

QSAR method and its application to toxicity prediction of metal nanomaterials / 中国药理学与毒理学杂志

Quantitative structure activity relationship ( QSAR ) study attempts to correlate chemical structure with activity using statistical approaches and is now being applied to high throughput toxicity screening and prediction of nanomaterials. This paper is interded to discuss the present QSAR study methods of nanomaterials based on traditional QSAR study, such as the use of measurement instrument and quantum chemistry methods of structure descriptor selection, evaluation criteria for the quality of published experimental data on nanomaterials, modeling methods such as K-nearest neighbor ( KNN) and support vector machine(SVM), validation methods such as leave-one-out(LOO) and leave-N-out ( LNO) . We also review the problems and challenges existing in this area and predict future development.

Predicting hepatotoxicity of compounds from traditional chinese medicines using tree models / 中国药学杂志

OBJECTIVE: To build tree models for the prediction of hepatotoxicity of compounds from traditional Chinese medicines (TCM).

Linear and non-linear quantitative structure-activity relationship models on indole substitution patterns as inhibitors of HIV-1 attachment.

The antiviral drugs that inhibit human immunodeficiency virus (HIV) entry to the target cells are already in different phases of clinical trials. They prevent viral entry and have a highly specific mechanism of action with a low toxicity profile. Few QSAR studies have been performed on this group of inhibitors. This study was performed to develop a quantitative structure–activity relationship (QSAR) model of the biological activity of indole glyoxamide derivatives as inhibitors of the interaction between HIV glycoprotein gp120 and host cell CD4 receptors. Forty different indole glyoxamide derivatives were selected as a sample set and geometrically optimized using Gaussian 98W. Different combinations of multiple linear regression (MLR), genetic algorithms (GA) and artificial neural networks (ANN) were then utilized to construct the QSAR models. These models were also utilized to select the most efficient subsets of descriptors in a cross-validation procedure for non-linear log (1/EC50) prediction. The results that were obtained using GA-ANN were compared with MLR-MLR and MLR-ANN models. A high predictive ability was observed for the MLR, MLR-ANN and GA-ANN models, with root mean sum square errors (RMSE) of 0.99, 0.91 and 0.67, respectively (N = 40). In summary, machine learning methods were highly effective in designing QSAR models when compared to statistical method.

A QSAR study on a series of simplified digitalis-like compounds acting on Na+,K+-ATPase

Among the cardiotonics (agents against congestive heart failure), the most important group is of the digitalis cardiac glycosides, but since these compounds suffer from a low therapeutic index, attention has been paid to investigating safer cardiotonic agents through the inhibition of Na+,K+-ATPase, the mechanism by which the digitalis cardiac glycosides elicit their action. Recently, a series of perhydroindenes were studied for their Na+,K+-ATPase inhibition activity. We report here a QSAR study on them to investigate the physicochemical and structural properties of the molecules that govern their activity in order to rationalize the structural modification to have more potent drugs. A multiple regression analysis reveals a significant correlation between the Na+,K+-ATPase inhibition activity of the compounds and Kier’s first order valence molecular connectivity index of their R5-substituents and some indicator parameters, suggesting that the R5-substituents of the compounds containing atoms with low valence and high saturation and the R1-substituents having =N−O− moiety will be conducive to the activity.