An improved ensemble learning machine for biological activity prediction of tyrosine kinase inhibitors

Boosting is one of the most important strategies in ensemble learning because of its ability to improve the stability and performance of weak learners. It is nonparametric, multivariate, fast and interpretable but is not robust against outliers. To enhance its prediction accuracy as well as immunize it against outliers, a modified version of a boosting algorithm (AdaBoost R2) was developed and called AdaBoost R3. In the sampling step, extremum samples were added to the boosting set. In the robustness step, a modified Huber loss function was applied to overcome the outlier problem. In the output step, a deterministic threshold was used to guarantee that bad predictions do not participate in the final output. The performance of the modified algorithm was investigated with two anticancer data sets of tyrosine kinase inhibitors, and the mechanism of inhibition was studied using the relative weighted variable importance procedure. Investigating the effect of base learner's strength reveals that boosting is only successful using the classification and regression tree method (a weak to moderate learner) and does not have a significant effect using the radial basis functions partial least square method (a strong base learners). Copyright © 2015 John Wiley & Sons, Ltd.     

CONFORMATIONAL PREFERENCES ABOUT Cpy-S BONDS IN DI-2-PYRIDYL DISULFIDE

Abstract

The total energy, dipole moment and electron densities for each possible rotational conformation about the C_py-S bonds of di-2-pyridyl disulfide were evaluated by using the semi-quantitative CNDO/2 method. The conformations in which the pyridine rings are coplanar with the valency plane of the bonded sulfur atom (cis-cis, cis-trans and trans-trans) were predicted to be the most favored ones.

Results of the theoretical study, when compared to some experimental determinations such as dipole moment and variable temperature pmr spectra, provided evidence that easy interconversion between these conformations can occur.     

CoMFA and CoMSIA studies on C-aryl glucoside SGLT2 inhibitors as potential anti-diabetic agents

SGLT2 has become a target of therapeutic interest in diabetes research. CoMFA and CoMSIA studies were performed on C-aryl glucoside SGLT2 inhibitors (180 analogues) as potential anti-diabetic agents. Three different alignment strategies were used for the compounds. The best CoMFA and CoMSIA models were obtained by means of Distill rigid body alignment of training and test sets, and found statistically significant with cross-validated coefficients (q ²) of 0.602 and 0.618, respectively, and conventional coefficients (r ²) of 0.905 and 0.902, respectively. Both models were validated by a test set of 36 compounds giving satisfactory predicted correlation coefficients (r ² _pred) of 0.622 and 0.584 for CoMFA and CoMSIA models, respectively. A comparison was made with earlier 3D QSAR study on SGLT2 inhibitors, which shows that our 3D QSAR models are better than earlier models to predict good inhibitory activity. CoMFA and CoMSIA models generated in this work can provide useful information to design new compounds and helped in prediction of activity prior to synthesis.     

靶向抑制拓扑异构酶和端粒酶的钌配合物研究进展

金属配合物抗肿瘤研究,尤其是铂类药物,已取得了相对令人瞩目的成功,但同时也面临着包括耐药性和毒副作用等诸多问题。近年来钌配合物作为新的抗癌药物引起了人们的注意。在非铂系药物中,金钌配合物;拓扑异构酶;G-四链体;端粒酶属钌配合物是最有前途的抗癌药物之一,国际上普遍认为钌和钌的配合物属于低毒性,容易吸收并在体内很快排泄。本文将着重介绍钌配合物与DNA结合后进一步引发的细胞内核酶活性抑制研究,从新的角度来诠释钌配合物的抗肿瘤研究最新进展。     

Screening of aromatase inhibitors in traditional Chinese medicines by electrophoretically mediated microanalysis in a partially filled capillary

An electrophoretically mediated microanalysis method with partial filling technique was developed for screening aromatase inhibitors in traditional Chinese medicine. The in‐capillary enzymatic reaction was performed in 20 mM sodium phosphate buffer (pH 7.4), and sodium phosphate buffer (20 mM, pH 8.0) was used as a background electrolyte. A long plug of coenzyme reduced β‐nicotinamide adenine dinucleotide 2′‐phosphate hydrate dissolved in the reaction buffer was hydrodynamically injected into a fused silica capillary followed by the injection of reaction buffer, enzyme, and substrate solution. The reaction was initiated with a voltage of 5 kV applied to the capillary for 40 s. The voltage was turned off for 20 min to increase the product amount and again turned on at a constant voltage of 20 kV to separate all the components. Direct detection was performed at 260 nm. The enzyme activity was directly assayed by measuring the peak area of the produced β‐nicotinamide adenine dinucleotide phosphate and the decreased peak area indicated the aromatase inhibition. Using the Lineweaver–Burk equation, the Michaelis–Menten constant was calculated to be 50 ± 4.5 nM. The method was applied to the screening of aromatase inhibitors from 15 natural products. Seven compounds were found to have potent AR inhibitory activity.     

Xanthione: A new and effective corrosion inhibitor for mild steel in sulphuric acid solution

The adsorption and inhibition effect of xanthione (XION) on mild steel in 0.5 M H₂SO₄ at 303–333 K were studied using gravimetric and UV–visible spectrophotometric methods. The results obtained show that XION acts as an effective corrosion inhibitor for mild steel in sulphuric acid and inhibition efficiency reaches 98.0% at a very low inhibitor concentration of 10 μM. Inhibition efficiency was found to increase with increase in XION concentration but decreased with temperature suggesting physical adsorption mechanism. Arrhenius law and its transition equation lead to estimate the activation parameters of the corrosion process. XION inhibits the corrosion of mild steel effectively at moderate temperature and adsorbs according to the Langmuir isotherm. Thermodynamic parameters governing the adsorption process have been calculated and discussed. The UV–visible absorption spectra of the solution containing the inhibitor after the immersion of mild steel specimen indicate the formation of a XEN–Fe complex. Attempt to correlate the molecular structure to quantum chemical indices was made using density functional theory (DFT).