Prediction of cytotoxicity data (CC(50)) of anti-HIV 5-phenyl-1-phenylamino-1H-imidazole derivatives by artificial neural network trained with Levenberg-Marquardt algorithm

Molecular docking and 4D-QSAR studies of metastatic cancer inhibitor thiazoles

By using the molecular docking and 4D-QSAR analysis, it is aimed to find the interaction points in the receptor binding site of transforming growth factor-beta (TGF-beta) used to inhibit invasion and metastasis. To elucidate the interaction points of receptor, different types of local reactive descriptor (LRD) of ligands have been used. Activity values related to interaction energy between the ligand-receptor (L-R) were determined by nonlinear least squares (NLLS) using the Levenberg-Marquardt (LM) algorithm. Using the Molecule Comparative Electron Topology (MCET) method, the 3D pharmacophore model (3D-PhaM) was obtained after alignment and superimposition of the molecules, and also confirmed by molecular docking method. With the leave one out-cross validation (LOO-CV) method, the best predictions are q² or r_CV² = 0.789 for the 51 compounds in the internal training set and r² = 0.785 for the 13 compounds in the external test set. Furthermore, the predictive capability of the advanced QSAR model is more precisely calculated with the r_m² metric (r_m² = 0.769).

Degradation and mineralization of phenol compounds with goethite catalyst and mineralization prediction using artificial intelligence

The efficiency of phenol degradation via Fenton reaction using mixture of heterogeneous goethite catalyst with homogeneous ferrous ion was analyzed as a function of three independent variables, initial concentration of phenol (60 to 100 mg /L), weight ratio of initial concentration of phenol to that of H₂O₂ (1: 6 to 1: 14) and, weight ratio of initial concentration of goethite catalyst to that of H₂O₂ (1: 0.3 to 1: 0.7). More than 90 % of phenol removal and more than 40% of TOC removal were achieved within 60 minutes of reaction. Two separate models were developed using artificial neural networks to predict degradation percentage by a combination of Fe³⁺ and Fe²⁺ catalyst. Five operational parameters were employed as inputs while phenol degradation and TOC removal were considered as outputs of the developed models. Satisfactory agreement was observed between testing data and the predicted values (R²_Phenol = 0.9214 and R²TOC= 0.9082).

2D-QSAR study of some 2,5-diaminobenzophenone farnesyltransferase inhibitors by different chemometric methods

Quantitative structure activity relationship (QSAR) models can be used to predict the activity of new drug candidates in early stages of drug discovery. In the present study, the information of the ninety two 2,5-diaminobenzophenone-containing farnesyltranaferase inhibitors (FTIs) were taken from the literature. Subsequently, the structures of the molecules were optimized using Hyperchem software and molecular descriptors were obtained using Dragon software. The most suitable descriptors were selected using genetic algorithms-partial least squares and stepwise regression, where exhibited that the volume, shape and polarity of the FTIs are important for their activities. The two-dimensional QSAR models (2D-QSAR) were obtained using both linear methods (multiple linear regression) and non-linear methods (artificial neural networks and support vector machines). The proposed QSAR models were validated using internal validation method. The results showed that the proposed 2D-QSAR models were valid and they can be used for prediction of the activities of the 2,5-diaminobenzophenone-containing FTIs. In conclusion, the 2D-QSAR models (both linear and non-linear) showed good prediction capability and the non-linear models were exhibited more accuracy than the linear models.

3D-MoRSE descriptors explained

3D-MoRSE is a very flexible 3D structure encoding framework for chemoinformatics and QSAR purposes due to the range of scattering parameter values and variety of weighting schemes used. While arising in many QSAR studies, up to this time they were considered as hardly interpreted and were treated like a "black box". This study is intended to lift the veil of mystery, providing a comprehensible way to the interpretation of 3D-MoRSE descriptors in QSAR/QSPR studies. The values of these descriptors are calculated with rather simple equation, but may vary when using differing starting geometries as optimization input. This variation increases with scattering parameter and also is higher for electronegativity weighted and unweighted descriptors. Though each 3D-MoRSE descriptor incorporates the information about the whole molecule structure, its final value is derived mostly from short-distance (up to 3Å) atomic pairs. And, if a QSAR study covers structurally similar set of compounds, then the role of 3D-MoRSE descriptor in a model can be interpreted using just several pairs of neighbor atoms. The guide to interpretation process is discussed and illustrated with a case study. Realizing the mathematical concept behind 3D-descriptors and knowing their properties it is easy not only to interpret, but also to predict the importance of 3D-MoRSE descriptors in a QSAR study. The process of prediction is described on the practical example and its accuracy is confirmed with further QSAR modeling.

Modeling and prediction of cytotoxicity of artemisinin for treatment of the breast cancer by using artificial neural networks

While artemisinin is known as anticancer medication with favorable remedial effects, its side effects must not be neglected. In order to reduce such side effects and increase artemisinin therapeutic index, nano technology has been considered as a new approach. Liposome preparation is supposed to be one of the new methods of drug delivery. To prepare the desired nanoliposome, certain proportions of phosphatidylcholine, cholesterol and artemisinin are mixed together. Besides, in order to achieve more stability, the formulation was pegylated by polyethylene glycol 2000 (PEG 2000). Mean diameter of nanoliposomes was determined by means of Zeta sizer. Encapsulation was calculated 96.02% in nanoliposomal and 91.62% in pegylated formulation. Compared to pegylated formulation, the percent of released drug in nanoliposomal formulation was more. In addition, this study reveals that cytotoxicity effect of pegylated nanoliposomal artemisinin was more than nanoliposomal artemisinin. Since artificial neural network shows high possibility of nonlinear modulation, it is used to predict cytotoxicity effect in this study, which can precisely indicate the cytotoxicity and IC50 of anticancer drugs.

Chemometrics analysis for investigation of retention behavior of hazardous compounds in effluents

The toxic substances, pesticides, and organic contaminants in effluents can potentially be causing damage that includes increased cancer risk; liver, kidney, stomach, nervous system, and immune system problems; reproductive difficulties; cataracts; and anemia. A quantitative structure-retention relationship (QSRR) was developed using the partial least square (PLS), kernel PLS (KPLS), and Levenberg-Marquardt artificial neural network (L-M ANN) approach for chemometrics study. The data which contained retention time (RT) of the 47 hazardous compounds in effluents were obtained by reverse-phase high-performance liquid chromatography. Genetic algorithm was employed as a factor selection procedure for PLS and KPLS modeling methods. By comparing the results, GA-PLS descriptors are selected for L-M ANN. Finally, a model with a low prediction error and a good correlation coefficient was obtained by L-M ANN. The described model does not require experimental parameters and potentially provides useful prediction for RT of new compounds. This is the first research on the QSRR of hazardous compounds in effluents using the chemometrics models.

Efficient approach for simultaneous estimation of multiple health-promoting effects of foods

The investigation of new food constituents for purposes of disease prevention or health promotion is an area of increasing interest in food science. This paper proposes a new system that allows for simultaneous estimation of the multiple health-promoting effects of food constituents using informatics. The model utilizes expression data of intracellular marker proteins as descriptors that reply to stimulation of a constituent. To estimate three health-promoting effects, namely, cancer cell growth suppression activity, antiviral activity, and antioxidant stress activity, each model was constructed using expression data of marker proteins as input data and health-promoting effects as the output value. When prediction performances of three types of mathematical models constructed by simple, multiple regressions, or artificial neural network (ANN), were compared, the most adequate model was the one constructed using an ANN. There were no statistically significant differences between the actual data and estimated values calculated by the ANN models. This system was able to simultaneously estimate health-promoting effects with reasonable precision from the same expression data of marker proteins. This novel system should prove to be an interesting platform for evaluation of the health-promoting effects of food.

Optimization of nonlinear dose- and concentration-response models utilizing evolutionary computation

An essential part of toxicity and chemical screening is assessing the concentrated related effects of a test article. Most often this concentration-response is a nonlinear, necessitating sophisticated regression methodologies. The parameters derived from curve fitting are essential in determining a test article's potency (EC(50)) and efficacy (E(max)) and variations in model fit may lead to different conclusions about an article's performance and safety. Previous approaches have leveraged advanced statistical and mathematical techniques to implement nonlinear least squares (NLS) for obtaining the parameters defining such a curve. These approaches, while mathematically rigorous, suffer from initial value sensitivity, computational intensity, and rely on complex and intricate computational and numerical techniques. However if there is a known mathematical model that can reliably predict the data, then nonlinear regression may be equally viewed as parameter optimization. In this context, one may utilize proven techniques from machine learning, such as evolutionary algorithms, which are robust, powerful, and require far less computational framework to optimize the defining parameters. In the current study we present a new method that uses such techniques, Evolutionary Algorithm Dose Response Modeling (EADRM), and demonstrate its effectiveness compared to more conventional methods on both real and simulated data.

Performance evaluation of radiologists with artificial neural network for differential diagnosis of intra-axial cerebral tumors on MR images

BACKGROUND AND PURPOSE

Previous studies have suggested that use of an artificial neural network (ANN) system is beneficial for radiological diagnosis. Our purposes in this study were to construct an ANN for the differential diagnosis of intra-axial cerebral tumors on MR images and to evaluate the effect of ANN outputs on radiologists' diagnostic performance.

MATERIALS AND METHODS

We collected MR images of 126 patients with intra-axial cerebral tumors (58 high-grade gliomas, 37 low-grade gliomas, 19 metastatic tumors, and 12 malignant lymphomas). We constructed a single 3-layer feed-forward ANN with a Levenberg-Marquardt algorithm. The ANN was designed to differentiate among 4 categories of tumors (high-grade gliomas, low-grade gliomas, metastases, and malignant lymphomas) with use of 2 clinical parameters and 13 radiologic findings in MR images. Subjective ratings for the 13 radiologic findings were provided independently by 2 attending radiologists. All 126 cases were used for training and testing of the ANN based on a leave-one-out-by-case method. In the observer test, MR images were viewed by 9 radiologists, first without and then with ANN outputs. Each radiologist's performance was evaluated through a receiver operating characteristic (ROC) analysis on a continuous rating scale.

RESULTS

The averaged area under the ROC curve for ANN alone was 0.949. The diagnostic performance of the 9 radiologists increased from 0.899 to 0.946 (P < .001) when they used ANN outputs.

CONCLUSIONS

The ANN can provide useful output as a second opinion to improve radiologists' diagnostic performance in the differential diagnosis of intra-axial cerebral tumors seen on MR imaging.