Design of N-11-Azaartemisinins Potentially Active against Plasmodium falciparum by Combined Molecular Electrostatic Potential, Ligand-Receptor Interaction and Models Built with Supervised Machine Learning Methods

N-11-azaartemisinins potentially active against Plasmodium falciparum are designed by combining molecular electrostatic potential (MEP), ligand-receptor interaction, and models built with supervised machine learning methods (PCA, HCA, KNN, SIMCA, and SDA). The optimization of molecular structures was performed using the B3LYP/6-31G* approach. MEP maps and ligand-receptor interactions were used to investigate key structural features required for biological activities and likely interactions between N-11-azaartemisinins and heme, respectively. The supervised machine learning methods allowed the separation of the investigated compounds into two classes: cha and cla, with the properties εLUMO+1 (one level above lowest unoccupied molecular orbital energy), d(C6-C5) (distance between C6 and C5 atoms in ligands), and TSA (total surface area) responsible for the classification. The insights extracted from the investigation developed and the chemical intuition enabled the design of sixteen new N-11-azaartemisinins (prediction set), moreover, models built with supervised machine learning methods were applied to this prediction set. The result of this application showed twelve new promising N-11-azaartemisinins for synthesis and biological evaluation.

Landslide susceptibility prediction using slope unit-based machine learning models considering the heterogeneity of conditioning factors

To perform landslide susceptibility prediction(LSP),it is important to select appropriate mapping unit and landslide-related conditioning factors.The efficient and automatic multi-scale segmentation(MSS)method proposed by the authors promotes the application of slope units.However,LSP modeling based on these slope units has not been performed.Moreover,the heterogeneity of conditioning factors in slope units is neglected,leading to incomplete input variables of LSP modeling.In this study,the slope units extracted by the MSS method are used to construct LSP modeling,and the heterogeneity of conditioning factors is represented by the internal variations of conditioning factors within slope unit using the descriptive statistics features of mean,standard deviation and range.Thus,slope units-based machine learning models considering internal variations of conditioning factors(variant slope-machine learning)are proposed.The Chongyi County is selected as the case study and is divided into 53,055 slope units.Fifteen original slope unit-based conditioning factors are expanded to 38 slope unit-based conditioning factors through considering their internal variations.Random forest(RF)and multi-layer perceptron(MLP)machine learning models are used to construct variant Slope-RF and Slope-MLP models.Meanwhile,the Slope-RF and Slope-MLP models without considering the internal variations of conditioning factors,and conventional grid units-based machine learning(Grid-RF and MLP)models are built for comparisons through the LSP performance assessments.Results show that the variant Slopemachine learning models have higher LSP performances than Slope-machine learning models;LSP results of variant Slope-machine learning models have stronger directivity and practical application than Grid-machine learning models.It is concluded that slope units extracted by MSS method can be appropriate for LSP modeling,and the heterogeneity of conditioning factors within slope units can more comprehensively reflect the relationships between conditioning factors and landslides.The research results have important reference significance for land use and landslide prevention.

A machine learning approach for accelerated design of magnesium alloys.Part B: Regression and property prediction

Machine learning(ML) models provide great opportunities to accelerate novel material development, offering a virtual alternative to laborious and resource-intensive empirical methods. In this work, the second of a two-part study, an ML approach is presented that offers accelerated digital design of Mg alloys. A systematic evaluation of four ML regression algorithms was explored to rationalise the complex relationships in Mg-alloy data and to capture the composition-processing-property patterns. Cross-validation and hold-out set validation techniques were utilised for unbiased estimation of model performance. Using atomic and thermodynamic properties of the alloys, feature augmentation was examined to define the most descriptive representation spaces for the alloy data. Additionally, a graphical user interface(GUI) webtool was developed to facilitate the use of the proposed models in predicting the mechanical properties of new Mg alloys. The results demonstrate that random forest regression model and neural network are robust models for predicting the ultimate tensile strength and ductility of Mg alloys, with accuracies of ~80% and 70% respectively. The developed models in this work are a step towards high-throughput screening of novel candidates for target mechanical properties and provide ML-guided alloy design.

The Coordinated Influence of Indian Ocean Sea Surface Temperature and Arctic Sea Ice on Anomalous Northeast China Cold Vortex Activities with Different Paths during Late Summer

The Northeast China cold vortex(NCCV)during late summer(from July to August)is identified and classified into three types in terms of its movement path using machine learning.The relationships of the three types of NCCV intensity with atmospheric circulations in late summer,the sea surface temperature(SST),and Arctic sea ice concentration(SIC)in the preceding months,are analyzed.The sensitivity tests by the Community Atmosphere Model version 5.3(CAM5.3)are used to verify the statistical results.The results show that the coordination pattern of East Asia-Pacific(EAP)and Lake Baikal high pressure forced by SST anomalies in the North Indian Ocean dipole mode(NIOD)during the preceding April and SIC anomalies in the Nansen Basin during the preceding June results in an intensity anomaly for the first type of NCCV.While the pattern of high pressure over the Urals and Okhotsk Sea and low pressure over Lake Baikal during late summer-which is forced by SST anomalies in the South Indian Ocean dipole mode(SIOD)in the preceding June and SIC anomalies in the Barents Sea in the preceding April-causes the intensity anomaly of the second type.The third type is atypical and is not analyzed in detail.Sensitivity tests,jointly forced by the SST and SIC in the preceding period,can well reproduce the observations.In contrast,the results forced separately by the SST and SIC are poor,indicating that the NCCV during late summer is likely influenced by the coordinated effects of both SST and SIC in the preceding months.

A Contemporary Review on Drought Modeling Using Machine Learning Approaches

Drought is the least understood natural disaster due to the complex relationship of multiple contributory factors. Itsbeginning and end are hard to gauge, and they can last for months or even for years. India has faced many droughtsin the last few decades. Predicting future droughts is vital for framing drought management plans to sustainnatural resources. The data-driven modelling for forecasting the metrological time series prediction is becomingmore powerful and flexible with computational intelligence techniques. Machine learning (ML) techniques havedemonstrated success in the drought prediction process and are becoming popular to predict the weather, especiallythe minimum temperature using backpropagation algorithms. The favourite ML techniques for weather forecastinginclude support vector machines (SVM), support vector regression, random forest, decision tree, logistic regression,Naive Bayes, linear regression, gradient boosting tree, k-nearest neighbours (KNN), the adaptive neuro-fuzzyinference system, the feed-forward neural networks, Markovian chain, Bayesian network, hidden Markov models,and autoregressive moving averages, evolutionary algorithms, deep learning and many more. This paper presentsa recent review of the literature using ML in drought prediction, the drought indices, dataset, and performancemetrics.

Real-time determination of sandy soil stiffness during vibratory compaction incorporating machine learning method for intelligent compaction

An emerging real-time ground compaction and quality control, known as intelligent compaction(IC), has been applied for efficiently optimising the full-area compaction. Although IC technology can provide real-time assessment of uniformity of the compacted area, accurate determination of the soil stiffness required for quality control and design remains challenging. In this paper, a novel and advanced numerical model simulating the interaction of vibratory drum and soil beneath is developed. The model is capable of evaluating the nonlinear behaviour of underlying soil subjected to dynamic loading by capturing the variations of damping with the cyclic shear strains and degradation of soil modulus. The interaction of the drum and the soil is simulated via the finite element method to develop a comprehensive dataset capturing the dynamic responses of the drum and the soil. Indeed, more than a thousand three-dimensional(3D) numerical models covering various soil characteristics, roller weights, vibration amplitudes and frequencies were adopted. The developed dataset is then used to train the inverse solver using an innovative machine learning approach, i.e. the extended support vector regression, to simulate the stiffness of the compacted soil by adopting drum acceleration records. Furthermore, the impacts of the amplitude and frequency of the vibration on the level of underlying soil compaction are discussed.The proposed machine learning approach is promising for real-time extraction of actual soil stiffness during compaction. Results of the study can be employed by practising engineers to interpret roller drum acceleration data to estimate the level of compaction and ground stiffness during compaction.

Prediction of Seaward Slope Recession in Berm Breakwaters Using M5＇ Machine Learning Approach

In the design process of berm breakwaters, their front slope recession has an inevitable rule in large number of model tests, and this parameter being studied. This research draws its data from Moghim's and Shekari's experiment results. These experiments consist of two different 2D model tests in two wave flumes, in which the berm recession to different sea state and structural parameters have been studied. Irregular waves with a JONSWAP spectrum were used in both test series. A total of 412 test results were used to cover the impact of sea state conditions such as wave height, wave period, storm duration and water depth at the toe of the structure, and structural parameters such as berm elevation from still water level, berm width and stone diameter on berm recession parameters. In this paper, a new set of equations for berm recession is derived using the M5' model tree as a machine learning approach. A comparison is made between the estimations by the new formula and the formulae recently given by other researchers to show the preference of new M5' approach.

基于监督下降法的大地电磁二维反演及应用研究

传统的大地电磁二维反演方法较为成熟,但仍存在反演结果依赖初始模型和正则化参数选取、容易陷入局部极小值等问题。监督下降法是一种学习平均下降方向来预测数据残差的机器学习算法。本文尝试采用监督下降法解决传统的大地电磁二维反演存在的问题,基于监督下降法理论开发了大地电磁二维反演算法,设计理论模型合成算例验证了算法的正确性,并对西藏高原实测数据进行反演,检验了监督下降法的实用性。理论模型合成数据和实测数据反演结果表明,相较于传统的非线性共轭梯度反演,基于监督下降法的反演具有收敛速度快、反演效果好、抗噪能力强等特点。

信息量法耦合机器学习模型的西山煤田滑坡易发性评价

采煤活动形成的地下采空区极易引发地质灾害,滑坡易发性评价是地质灾害风险预警的先行工作。以山西省西山煤田为研究区,构建了20个滑坡致灾因子,利用信息量(Information Value,IV)法耦合逻辑回归(Logistic Regression,LR)、随机森林(Random Forest,RF)和支持向量机(Support Vector Machine,SVM)模型,构建IV-LR、IV-RF和IV-SVM这3种IV法耦合机器学习模型,并进行研究区滑坡易发性评价,通过受试者工作特征(Receiver Operating Characteristic,ROC)曲线、均值和标准差分析建模结果精度。结果表明,研究区内高、极高易发区主要分布在距水系300 m内,极低、低易发区分布在中西部地区,IV-LR、IV-RF和IV-SVM模型验证精度分别为76.67%、74.62%和78.57%,ROC曲线下面积(Area Under Curve,AUC)为0.86、0.83和0.84。IV-LR模型AUC值最大,预测精度最高。

基于层一致性平均教师模型的半监督岩石薄片图像分类

传统的岩石薄片图像分类依赖于大量人工标记的图像样本,这种方式受制于标记人员的经验和能力,且无法通过不断增加的未标记岩石薄片图像样本实现分类能力的可扩展式增强。该文提出的在平均教师(mean teacher, MT)模型的基础上,通过在无监督损失中添加层一致性正则化项的方式约束师生模型的层次结构,实现对未标记数据信息的有效利用。消融实验和层一致性平均教师(hierarchy consistency mean teacher, HCMT)模型对比实验结果表明,层一致性正则化方法利用了未标记数据的有效信息,提升了MT模型的分类效果,使得HCMT模型可以在半标记数据集中获得如全标记数据集相似的分类能力。该实验表明,半监督学习模型利用大量未标记岩石薄片图像数据可以提升模型分类的能力。

新发展格局提质之谜:从全面创新改革到区域创新生态系统演化

本文基于国内循环质量和国际循环质量两个维度评价了新发展格局质量,并从繁衍、变异、适应、流动、锁定破解5个维度评价了区域创新生态系统演化指数,形成2010~2022年30个省(区、市)的面板数据,采用空间双重差分和双重机器学习模型对“全面创新改革试验-区域创新生态系统演化-新发展格局质量”研究系统进行准自然实验,得到结论如下:(1)全面创新改革试验和区域创新生态系统向更高阶段演化均能够显著促进新发展格局质量;(2)在其他地区进行全面创新改革试验能够对本地区新发展格局质量具有促进效应,但区域创新生态系统向更高阶段演化对新发展格局质量的空间传导效应显著为负;(3)全面创新改革试验能够通过正向促进区域创新生态系统演化对区域国内循环质量和国际循环质量产生正向影响。

Uncertainties of landslide susceptibility prediction: Influences of random errors in landslide conditioning factors and errors reduction by low pass filter method

In the existing landslide susceptibility prediction(LSP)models,the influences of random errors in landslide conditioning factors on LSP are not considered,instead the original conditioning factors are directly taken as the model inputs,which brings uncertainties to LSP results.This study aims to reveal the influence rules of the different proportional random errors in conditioning factors on the LSP un-certainties,and further explore a method which can effectively reduce the random errors in conditioning factors.The original conditioning factors are firstly used to construct original factors-based LSP models,and then different random errors of 5%,10%,15% and 20%are added to these original factors for con-structing relevant errors-based LSP models.Secondly,low-pass filter-based LSP models are constructed by eliminating the random errors using low-pass filter method.Thirdly,the Ruijin County of China with 370 landslides and 16 conditioning factors are used as study case.Three typical machine learning models,i.e.multilayer perceptron(MLP),support vector machine(SVM)and random forest(RF),are selected as LSP models.Finally,the LSP uncertainties are discussed and results show that:(1)The low-pass filter can effectively reduce the random errors in conditioning factors to decrease the LSP uncertainties.(2)With the proportions of random errors increasing from 5%to 20%,the LSP uncertainty increases continuously.(3)The original factors-based models are feasible for LSP in the absence of more accurate conditioning factors.(4)The influence degrees of two uncertainty issues,machine learning models and different proportions of random errors,on the LSP modeling are large and basically the same.(5)The Shapley values effectively explain the internal mechanism of machine learning model predicting landslide sus-ceptibility.In conclusion,greater proportion of random errors in conditioning factors results in higher LSP uncertainty,and low-pass filter can effectively reduce these random errors.

Tunable structure priors for Bayesian rule learning for knowledge integrated biomarker discovery

AIM To develop a framework to incorporate background domain knowledge into classification rule learning for knowledge discovery in biomedicine.METHODS Bayesian rule learning(BRL) is a rule-based classifier that uses a greedy best-first search over a space of Bayesian belief-networks(BN) to find the optimal BN to explain the input dataset, and then infers classification rules from this BN. BRL uses a Bayesian score to evaluate the quality of BNs. In this paper, we extended the Bayesian score to include informative structure priors, which encodes our prior domain knowledge about the dataset. We call this extension of BRL as BRL_p. The structure prior has a λ hyperparameter that allows the user to tune the degree of incorporation of the prior knowledge in the model learning process. We studied the effect of λ on model learning using a simulated dataset and a real-world lung cancer prognostic biomarker dataset, by measuring the degree of incorporation of our specified prior knowledge. We also monitored its effect on the model predictive performance. Finally, we compared BRL_p to other stateof-the-art classifiers commonly used in biomedicine.RESULTS We evaluated the degree of incorporation of prior knowledge into BRL_p, with simulated data by measuring the Graph Edit Distance between the true datagenerating model and the model learned by BRL_p. We specified the true model using informative structurepriors. We observed that by increasing the value of λ we were able to increase the influence of the specified structure priors on model learning. A large value of λ of BRL_p caused it to return the true model. This also led to a gain in predictive performance measured by area under the receiver operator characteristic curve(AUC). We then obtained a publicly available real-world lung cancer prognostic biomarker dataset and specified a known biomarker from literature [the epidermal growth factor receptor(EGFR) gene]. We again observed that larger values of λ led to an increased incorporation of EGFR into the final BRL_p model. This relevant background knowledge also led to a gain in AUC.CONCLUSION BRL_p enables tunable structure priors to be incorporated during Bayesian classification rule learning that integrates data and knowledge as demonstrated using lung cancer biomarker data.

基于半监督高斯混合模型与梯度提升树的砂岩储层相控孔隙度预测

孔隙度是一种描述储层物性特征的重要参数。考虑砂岩与泥岩的孔隙度存在明显差异,提出了一种基于半监督高斯混合模型与梯度提升树的相控孔隙度预测方法,以实现砂岩储层孔隙度的精细描述。首先利用少量具岩相标签的测井数据确定高斯混合模型的初始聚类中心及对应的岩相类别;其次利用大量无标签测井数据优化高斯混合模型,实现砂岩与泥岩的准确划分;再次基于地质认识将泥岩孔隙度解释为固定的极小值,从而后续只开展砂岩孔隙度预测;然后将测井曲线拟合方法导出的孔隙度先验信息和测井敏感属性作为梯度提升树算法的多元输入信息,通过学习统计性岩石物理关系建立砂岩孔隙度的计算模型;最终根据岩相结果将砂岩段与泥岩段的孔隙度进行组合得到相控孔隙度。D油田的18口井数据测试结果表明:半监督高斯混合模型的岩相分类效果优于K均值、支持向量机、随机森林等机器学习算法,2口盲井的岩相分类准确率达到94.5%;所构建方法对2口盲井预测的相控孔隙度结果与真实孔隙度具有较高的一致性,平均相关系数达0.805。

建成环境对城市轨道交通起讫点客流的非线性影响及阈值效应

城市轨道交通起讫点(OD)客流与建成环境的依赖关系研究有助于TOD(Transit-oriented development)模式的实施。既有研究多关注建成环境对进出站客流的影响,而基于OD客流的研究未充分考虑建成环境要素的交互效应对OD客流的影响。采用多源位置大数据系统地刻画城市轨道交通的TOD建成环境,基于极限梯度提升决策树模型(XGBoost)研究城市轨道交通OD客流与TOD建成环境的非线性关系。针对北京地铁的案例研究表明:XGBoost能有效地处理建成环境对OD客流的非线性影响,其解释能力达到72.6%,估计结果更为可靠。TOD建成环境因子对OD客流的影响差异显著。密度和公共交通可达性等两类要素的重要度排序前二,其解释变量的平均重要度达到4.41%和3.71%,是全部变量平均值的1.29倍和1.08倍。解释变量重要度排序高的建成环境因子对OD客流的影响存在非线性特征,表现为显著的阈值效应。基于双变量部分依赖图的分析表明,城市轨道交通客流的流动依赖于起讫点建成环境的差异及其引发的交互效应。因此,发展城市轨道交通TOD时,不仅需从交通生成角度分析建成环境对进出站客流的影响,还需考虑客流的矢量性,从交通分布角度研究建成环境各要素的资源协同配置问题。

基于SGMD及LWOA-ELM的有限元模型修正

为得到待修正参数与结构响应之间的关系,提高模型修正的效率和精度,提出了一种基于辛几何模态分解(SGMD)和Lévy飞行鲸鱼优化算法(LWOA)优化极限学习机(ELM)的有限元模型修正(FEMU)方法。首先,对加速度频响函数(AFRF)进行SGMD分解,采用能量熵增量法确定重组辛几何分量(SGC)构成SGC矩阵。然后,利用LWOA对ELM的权值和阈值进行优化,提高ELM模型的预测效率,以LWOA-ELM为代理模型映射出待修正参数与SGC矩阵之间的关系。最后,以试验频响函数SGC矩阵与LWOA-ELM模型输出所得矩阵差值的F-范数最小为目标函数,结合LWOA求解待修正参数。算例分析表明,提出的方法用于有限元模型修正有较好的可行性和有效性。以SGC矩阵表征AFRF的修正方法,有较好的噪声鲁棒性;LWOA-ELM作为代理模型预测精度高,泛化能力强。

竖缝式鱼道内齐口裂腹鱼洄游行为模拟

鱼道是帮助洄游鱼类跨越河道内物理障碍的主要过鱼设施之一。针对鱼道结构优化试验耗时长、成本高的问题。该研究在物理试验基础上建立模拟鱼类运动的数值模型,基于竖缝式鱼道内齐口裂腹鱼的行为数据,结合机器学习和欧拉-拉格朗日智能体方法(Eulerian-Lagrangian-agent method,ELAM)建立了鱼类洄游路线预测模型。首先,将齐口裂腹鱼的游泳行为数据划分为3个数据集(顺流而下、侧向运动和逆流而上);然后,根据不同机器学习算法(XGBoost、K-近邻、梯度提升决策树、随机森林)分别构建游泳行为分类模型和游泳速度回归模型,并通过试验结果对比,筛选出最优模型;最后,结合ELAM的框架,构建鱼类洄游数学模型,模拟了齐口裂腹鱼在3种体型的竖缝式鱼道中的洄游路线。结果表明:基于随机森林建立的游泳行为分类模型和游泳速度回归模型预测效果最佳,游泳行为分类精度为0.804,召回率为0.794,F1得分为0.798,3个数据集的游泳速度回归的R2均大于0.75。研究所建立的鱼类洄游数学模型能够较好的预测齐口裂腹鱼的洄游路线,可为相关鱼类保护措施的设计和优化提供参考。

监督学习在预测危险因素诱导下急性呼吸窘迫综合征发生风险中的应用

急性呼吸窘迫综合征(acute respiratory distress syndrome,ARDS)缺乏特异性诊断标准,且诱因复杂,在临床实践中往往难以做到早期识别、及时干预,这就需要一种精确、高效的手段辅助识别其发生。基于大数据的机器学习作为一种可以处理海量数据、高效利用有效知识的学习方法,在众多领域发挥了不同作用,在医学领域的重要性日益凸显。截至目前,在医学领域已有大量的机器学习成功应用的案例,其中监督学习算法凭借其可以预测风险的优势,获得众多研究者青睐。本文旨在阐述机器学习算法中监督学习算法在预测危险因素诱导下ARDS发生风险的临床应用。

基于随机IDA和机器学习的盾构隧道地震易损性分析

在隧道地震易损性分析中,涉及地震动、结构、岩土参数的随机性和不确定性。当考虑多种参数不确定性时,基于数值法建立结构地震易损性曲线需要进行大量工况的随机地震响应计算,面临很高的计算成本。为兼顾隧道地震易损性结果的客观性和计算的高效性,基于Python语言编写自动随机IDA脚本程序,实现隧道随机地震响应的自动分析和后处理功能。在此基础上,结合10种机器学习算法和地震易损性分析理论,提出了基于随机IDA和机器学习的盾构隧道地震易损性分析框架。以某盾构隧道为例,在建立确定性动力响应计算模型基础上,采用拉丁超立方抽样方法并结合自动随机IDA程序,完成了隧道随机增量动力分析并形成机器学习算法的数据集。通过特征选择、数据集划分、预处理及参数调优,建立了预测结构地震损伤的10种机器学习算法模型,对比分析了不同机器学习算法的预测性能,进一步讨论了土层随机参数对结构易损性的敏感性。结果表明:各机器学习算法模型在数据集上均表现出较好的预测性能,其中BPNN模型的综合预测性能最好。基于机器学习模型预测建立的隧道易损性曲线,与数值法结果总体一致,说明采用机器学习算法预测隧道地震易损性是有效的,特别是BPNN算法模型,与数值法非常接近,体现较高的可靠性。对样本数量敏感性小的机器学习算法,如BPNN算法模型在易损性预测中具有很好的适用性和应用潜力。从土体物理力学参数对盾构隧道地震易损性影响的敏感性程度而言,由大至小依次为弹性模量、泊松比、密度、阻尼、摩擦角和黏聚力。

基于XGBoost的二轮车碾压事故致因研究

二轮车被卷入机动车底部并遭受碾压,会造成严重的事故伤害。为识别二轮车被卷入机动车底部的原因及其影响因素,以中国交通事故深度调查(CIDAS)数据库中2 627起二轮车与机动车碰撞事故案例为基础,根据数据分布特征采用合成少数过采样(SMOTE)技术形成平衡数据集以训练机器学习模型;经过对比7种机器学习模型的分类性能,选用XGBoost模型构建二轮车碾压事故预测模型,并基于多项指标进行交叉验证以验证模型的预测性能;最后采用SHAP可解释性方法进一步挖掘二轮车碾压事故致因。结果表明:二轮车碾压事故致死率超出非碾压事故25.3%;XGBoost模型的综合预测性能优于其余6种机器学习模型;在与事故相关的环境因素中,工业区和郊区以及弯道、交叉路口是二轮车碾压事故的高发地点;与碰撞场景相关的二轮车碾压事故高风险因素包括二轮车侧面或尾部碰撞以及二轮车较低车速;摩托车以及踏板式二轮车不易被四轮机动车碾压;此外,二轮车车身越小,四轮机动车越高、越长,越容易发生二轮车碾压事故。根据研究结果从车辆设计、交通管理角度提出了避免二轮车骑行者遭受四轮机动车碾压的建议,为事故精准防控提供了新的视角和信息。