This work reports the classification study conducted on the biggest COX-2 inhibitor data set so far. Using 2925 diverse COX-2 inhibitors collected from 168 pieces of literature, we applied machine learning methods, support vector machine (SVM) and random forest (RF), to develop 12 classification models. The best SVM and RF models resulted in MCC values of 0.73 and 0.72, respectively. The 2925 COX-2 inhibitors were reduced to a data set of 1630 molecules by removing intermediately active inhibitors, and 12 new classification models were constructed, yielding MCC values above 0.72. The best MCC value of the external test set was predicted to be 0.68 by the RF model using ECFP_4 fingerprints. Moreover, the 2925 COX-2 inhibitors were clustered into eight subsets, and the structural features of each subset were investigated. We identified substructures important for activity including halogen, carboxyl, sulfonamide, and methanesulfonyl groups, as well as the aromatic nitrogen atoms. The models developed in this study could serve as useful tools for compound screening prior to lab tests.

No abstract

No abstract

ABSTRACT. This paper discusses the notch effect on multiaxial low cycle fatigue. Neuber's rule was firstly introduced to estimate the local strain at the notch root in proportional tension and torsion loading. The Neuber's rule was applied to estimating crack initiation and for propagation lives in tension-torsion low cycle fatigue. The rule conservatively estimated the crack initiation life in tension low cycle fatigue and appropriately in torsion low cycle fatigue. A simple method for estimating the local strain at the notch root was proposed in tension and torsion loading. The notch effect in nonproportional low cycle fatigue was discussed in two materials. The local strain at the notch root obtained by finite element analysis underestimated the crack initiation lives for the additional hardening material but that obtained by the Neuber's rule overestimated for the non-additional hardening material.

No abstract

The radial velocity (RV) is a basic physical quantity which can be determined through Doppler shift of the spectrum of a star. The precision of RV measurement depends on the resolution of the spectrum we used and the accuracy of wavelength calibration. In this work, radial velocities of LAMOST-II medium resolution (R ∼ 7500) spectra are measured for 1,594,956 spectra (each spectrum has two wavebands) through matching with templates. A set of RV standard stars are used to recalibrate the zero point of the measurement, and some reference sets with RVs derived from medium/high-resolution observations are used to evaluate the accuracy of the measurement. Comparing with reference sets, the accuracy of our measurement can get 0.0227 km s −1 with respect to radial velocities standard stars. The intrinsic precision is estimated with the multiple observations of single stars, which can achieve to 1.36 km s −1 ,1.08 km s −1 , 0.91 km s −1 for the spectra at signal-to-noise levels of 10, 20, 50, respectively. These big samples of RVs were obtained through spectra with different resolving power, which provides astronomers useful tools for dissecting and understanding the structure of the Milky Way. A large sample with consistent measured RVs is a key underpinning for a lot of research work, for example, Geller et al. (2015) use RVs to determine membership of a stellar cluster. Besides, researchers analyzed the RV variations to constrain the stellar pulsation model (Britavskiy et al. 2018) and to determinate the properties of the eclipsing binary stars (He lminiak et al. 2018; Martin et al. 2019).

No abstract

No abstract