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.     

Induction of decision trees using genetic programming for modelling ecotoxicity data: adaptive discretization of real-valued endpoints

Recent literature has demonstrated the applicability of genetic programming to induction of decision trees for modelling toxicity endpoints. Compared with other decision tree induction techniques that are based upon recursive partitioning employing greedy searches to choose the best splitting attribute and value at each node that will necessarily miss regions of the search space, the genetic programming based approach can overcome the problem. However, the method still requires the discretization of the often continuous-valued toxicity endpoints prior to the tree induction. A novel extension of this method, YAdapt, is introduced in this work which models the original continuous endpoint by adaptively finding suitable ranges to describe the endpoints during the tree induction process, removing the need for discretization prior to tree induction and allowing the ordinal nature of the endpoint to be taken into account in the models built.     

大型砂尘风洞中砂尘浓度测量与控制方法的研究

砂尘浓度指标严重影响砂尘风洞试验效果;在大型砂尘风洞中准确的测量及有效的控制砂尘浓度是砂尘风洞设备研制中的两个关键问题;针对动态试验工况环境中浓度传感器标定的问题,分析了浓度测量时标定参数与风速和浓度的关系,使用模型树算法建立了三者之间的关系模型,以此模型来预测各种动态情况下的标定参数;针对复杂环境下浓度动态控制问题,采用专家PID算法,提出了多回路控制系统结构和专家规则;工程实践表明该测量和控制方法满足了动态试验环境的需求,取得了良好效果;模型树算法和专家PID系统能有效的解决动态工况下砂尘浓度的测量与控制问题。     

Asymptotic expansions of the moments of extremes from general error distribution

In this paper, we derive the asymptotic expansions of the moments of normalized partial maxima for general error distribution. A byproduct is to deduce the convergence rates of the moments of normalized maxima to the moments of the corresponding extreme value distribution.     

非齐次树上马氏链的若干强偏差定理

通过构造适当的非负鞅,将Doob鞅收敛定理应用于几乎处处收敛的研究,给出了一类非齐次树上马氏链场加权和滑动平均的若干强偏差定理.     

Uncommon suffix tries

Common assumptions on the source producing the words inserted in a suffix trie with n leaves lead to a height and saturation level. We provide an example of a suffix trie whose height increases faster than a power of n and another one whose saturation level is negligible with respect to . Both are built from VLMC (Variable Length Markov Chain) probabilistic sources and are easily extended to families of tries having the same properties. The first example corresponds to a “logarithmic infinite comb” and enjoys a non uniform polynomial mixing. The second one corresponds to a “factorial infinite comb” for which mixing is uniform and exponential. © 2013 Wiley Periodicals, Inc. Random Struct. Alg., 46, 117–141, 2015     

MFCAV近似Riemann解法器在相容拉氏方法中的熵条件分析

 对基于MFCAV(Multi Fluid Channel on Averaged Volume)近似Riemann解法器的相容拉氏方法的熵条件进行了分析. 结果表明与满足声学形式Riemann解法器的熵不同, 前者只能在每个网格边界左、右两侧网格的熵随时间变化的和保证大于零, 即能保证整体熵增, 但不保证传统意义上的在每个网格中的熵增;而后者不仅保证整体熵增, 而且还满足传统意义上的熵增. 因此MFCAV的熵增相对声学形式解法器而言要弱一些, 由此表明其熵增可能要小些, 使得格式的耗散可能要小些.数值算例也验证了分析的正确性.     

Covariate model-based fault tree analysis for risk assessment in chemical process industries: A case study of a chlorine manufacturing facility

Fault tree analysis (FTA) is a promising quantitative technique for risk analysis in chemical process industries (CPIs). In FTA, a certain sequence of basic events (causes) leads to one specific Top event (critical event of interest). However, the conventional fault tree analysis has the limitations of staticity and uncertainty. The staticity in conventional FTA arises due to its inability to accommodate time-dependent characteristics of the process system. Whereas uncertainty primarily lies in the failure probability data of basic events. This paper proposes an innovative methodology that uses a time-dependent covariate model to update the failure probability values of major contributing basic events in FTA. A novel subclass of the family of phase-type distributions is used to model the covariates corresponding to the basic events. The newly developed methodology is applied for a case study in a chlorine manufacturing facility to estimate the chlorine release probability. The blockage in the pipeline was identified as the significant reason for chlorine release from expert opinion and sensitivity analysis. The results of the proposed model of FTA are compared with that of conventional FTA.     

Development of GC–MS/MS method for environmental monitoring of 49 pesticide residues in food commodities in Al-Rass,Al-Qassim region,Saudi Arabia

The harmful effects of pesticide residues are a threat to our health. Therefore, the current study aimed to validate a simple method for the determination of pesticide residues in commonly consumed fruits and vegetables from Al-Rass, Al-Qassim region, Saudi Arabia. A total of 1430 samples were collected from a local market and then analyzed for monitoring of 49 pesticide residues. A quick, easy, cheap, effective, rugged, and safe (QuEChERS) multi-residue extraction method followed by gas chromatography equipped with triple-quadrupole mass spectrometry (GC–MS/MS) was successfully implemented. This 17-min-run analytical method detects and quantifies pesticide residues with acceptable validation performance parameters in terms of sensitivity, selectivity, linearity, the limit of quantification, accuracy, and precision. The linear range of the calibration curves ranged from 10 to 300 µg/L, all the pesticide LODs ranged from 0.0005 to 0.0024 mg/kg, and the pesticide LOQs ranged from 0.0011 to 0.0047 mg/kg. The recovery values at the three fortification levels ranged from 78 % to 107 %, and the precision values (expressed as RSD%) were less than 20 % for all of the investigated analytes. The results showed that 138 (9.65 %) of the analyzed samples were contaminated with pesticide residues, 40 (2.80 %) of the analyzed samples exceeded the maximum residue limit (MRL) of the European Commission regulations (EC) for pesticides residues, 98 (6.85 %) of the analyzed samples were contaminated with residues below the MRL, and 1292 (90.35 %) of the analyzed samples were pesticide residue-free. Coriander contained the highest percentage (46.88 %) of pesticide residues, particularly tetradifon that representing 18.75 % noncompliance with the MRL, followed by parsley, with 20.59 % pesticide residues (10.29 % non-compliance). Multiple pesticide residues were observed most frequently in tomatoes and dates which were contaminated with buprofezin and ethion respectively.     

Nonuniform sampling and maximum entropy reconstruction applied to the accurate measurement of residual dipolar couplings

Residual dipolar couplings (RDC) provide important global restraints for accurate structure determination by NMR. We show that nonuniform sampling in combination with maximum entropy reconstruction (MaxEnt) is a promising strategy for accelerating and potentially enhancing the acquisition of RDC spectra. Using MaxEnt-processed spectra of nonuniformly sampled data sets that are reduced up to one fifth relative to uniform sampling, accurate 13C'-13Calpha RDCs can be obtained that agree with an RMS of 0.67 Hz with those derived from uniformly sampled, Fourier transformed spectra. While confirming that frequency errors in MaxEnt spectra are very slight, an unexpected class of systematic errors was found to occur in the 6th significant figure of 13C' chemical shifts of doublets obtained by MaxEnt reconstruction. We show that this error stems from slight line shape perturbations and predict it should be encountered in other nonlinear spectral estimation algorithms. In the case of MaxEnt reconstruction, the error can easily be rendered systematic by straightforward optimization of MaxEnt reconstruction parameters and self-cancels in obtaining RDCs from nonuniformly sampled, MaxEnt reconstructed spectra.