4D-QSAR Analyses for EGFR Inhibitors Based on CDDA-OPS-GA Method

Epidermal growth factor receptor is a preferred target for treating cancer. Compared to 3D-QSAR, 4D-QSAR has the feature of conformational flexibility and free alignment for individual ligands. In present studies, the 4D-QSAR of 131 analogs of 4-anilino quinazoline for EGFR inhibitors was built. The GROMACS package was employed to yield the conformational ensemble profile. The field descriptors of Coulomb and Lennard−Jones potentials were calculated by LQTA-QSAR. The filter descriptors and variable selection is very important, which was performed by means of comparative distribution detection algorithm (CDDA), ordered predictors selection (OPS) and genetic algorithm (GA) method. Best 4D-QSAR model yielded satisfactory statistics (R2 = 0.71), good performance in internal (Q2LOO = 0.60) and external prediction (R2pred = 0.69, k = 0.97, k′ = 1.01). The 4D-QSAR was shown to be robust (Q2LMO = 0.59) and was not built by chance (R2YS = 0.17, Q2YS = −0.25). The model has a good potential for rational design new EGFR inhibitors.

Real-time liver tracking algorithm based on LSTM and SVR networks for use in surface-guided radiation therapy

Background Surface-guided radiation therapy can be used to continuously monitor a patient’s surface motions during radiotherapy by a non-irradiating, noninvasive optical surface imaging technique. In this study, machine learning methods were applied to predict external respiratory motion signals and predict internal liver motion in this therapeutic context. Methods Seven groups of interrelated external/internal respiratory liver motion samples lasting from 5 to 6 min collected simultaneously were used as a dataset, Dv. Long short-term memory (LSTM) and support vector regression (SVR) networks were then used to establish external respiratory signal prediction models (LSTMpred/SVRpred) and external/internal respiratory motion correlation models (LSTMcorr/SVRcorr). These external prediction and external/internal correlation models were then combined into an integrated model. Finally, the LSTMcorr model was used to perform five groups of model updating experiments to confirm the necessity of continuously updating the external/internal correlation model. The root-mean-square error (RMSE), mean absolute error (MAE), and maximum absolute error (MAX_AE) were used to evaluate the performance of each model. Results The models established using the LSTM neural network performed better than those established using the SVR network in the tasks of predicting external respiratory signals for latency-compensation (RMSE < 0.5 mm at a latency of 450 ms) and predicting internal liver motion using external signals (RMSE < 0.6 mm). The prediction errors of the integrated model (RMSE ≤ 1.0 mm) were slightly higher than those of the external prediction and external/internal correlation models. The RMSE/MAE of the fifth model update was approximately ten times smaller than that of the first model update. Conclusions The LSTM networks outperform SVR networks at predicting external respiratory signals and internal liver motion because of LSTM’s strong ability to deal with time-dependencies. The LSTM-based integrated model performs well at predicting liver motion from external respiratory signals with system latencies of up to 450 ms. It is necessary to update the external/internal correlation model continuously.

QSAR and Anticancer Drug Design on Benzothienopyrimidinones as Promising Pim Kinase Inhibitors Utilizing Structural Descriptors

Pim kinase is a major target of anticancer chemotherapeutics. There are a number of pim kinase inhibitors which are being under clinical trials. But there are only a few QSAR and drug design attempts targeting pim kinase inhibition reported as of yet. Several 3H-Benzo[4,5]thieno[3,2-d]pyrimidin-4-one derivatives are taken into consideration here for the development of QSAR models utilizing topological and three dimensional structural indices against pim-1 and pim-2 kinase. Interesting results were found out where a model can produce an external prediction of 62.2% and 58.4% of variances for the inhibition of pim 1 and pim 2 kinases of the benzothienopyrimidinone compounds. Validated models have captured important structural features including electronegativity, polarizability and radial distribution function necessary for the inhibition of pim kinase. Such models have been utilized to screen a number of congeneric compounds from external ChMBL database. Further molecular docking was done to predict the mode of which may help to design new active congeners which are potent to inhibit pim kinase.

THE RESEARCH INTO THE PROBLEM OF STATISTICALLY INDETERMINATE TIME SERIES PREDICTION

In the article under consideration the appropriateness of prediction task optimizing according to related external prediction quality requirements based on the multitude of elements expanded by identification methods is proved by 15 mathematical models of time series and 4 methods of their identification.

A QSAR Study of Human Thymidine Phosphorylase Inhibitors with SMILES-Based Descriptors

Thymidine phosphorylase (TP) is a multifunctional protein frequently overexpressed in many types of cancer. Considering the interest in new anticancer compounds, a QSAR study was carried out to investigate a set of uracil derivatives described as TP inhibitors. The only molecular descriptors used were derived from SMILES notation. Ordered Predictors Selection (OPS) was used for variable selection and the models were built using the PLS method. The authors validated the internal and external prediction capabilities of the obtained model. The model was also tested using a set of 12 molecules obtained by similarity search in the ZINC database. The results showed that it is possible to describe the variation of biological activity of the selected dataset using only SMILES-derived molecular descriptors, and the obtained model shows potential for use as an aid in the design of new TP inhibitors.

Prediction of external wave pressures on a rubble mound breakwater

The external pressures on the front face of a rubble mound breakwater resulting from wave attack are examined in this paper. This is done through extensive model tests employing regular waves up to 30 cm in height, on a conventionally designed breakwater with front slopes of 1:1.5, 1:2, and 1:3. The measured pressures are examined based on their relationship to a number of different parameters, including wave steepness, wave height, wave period, breakwater front slope, core permeability, and elevation on the breakwater relative to the still water level. The average differential pressure, the maximum recorded differential pressure, the average minimum pressure, and the pressure rise and fall times are investigated, producing a regression equation for each case based on a number of independent variables. The regression equations demonstrate the great effect of the elevation on the breakwater, and often wave steepness; the much lesser effect attributed to the breakwater front slope; and the minimal effect that the core permeability has on most of the components describing the external pressures measured on a breakwater under wave attack. Key words: breakwater, rubble, pressure, external, prediction.