基于网络药理学香青兰防治痴呆的机制探讨

目的 采用网络药理学技术探讨香青兰防治痴呆的物质基础和作用机制。方法 通过文献挖掘和药动学参数筛选香青兰含有的活性化合物,通过检索DrugBank、GeneCards和OMIM数据库收集与痴呆相关的靶标;运用Cytoscape软件构建香青兰防治痴呆的蛋白质相互作用网络;进行基因本体和信号通路富集分析;构建香青兰化合物-关键靶标-通路网络。结果 筛选得到香青兰中药物活性成分42种,获得香青兰防治痴呆的潜在药物靶标90个,于蛋白质互作网络中进一步筛选得到74个主要靶标,富集得到300条信号通路、1 465个生物学过程、104个细胞组分和108个分子功能,通过化合物-靶标-通路网络中获得香青兰发挥防治痴呆作用的关键化合物,如金合欢素、芹菜素、金圣草（黄）素等黄酮类化合物,以及对应的关键靶标与关键信号通路,如MAPK、APP、MAPT、PI3K等靶标与阿尔茨海默病、神经营养蛋白等通路。结论 香青兰防治痴呆具有多成分、多靶标、多通路的作用特点。     

基于近红外和化学计量学方法对金线莲快速鉴别及其多糖含量测定的研究

目的 建立近红外分析方法结合化学计量学算法对金线莲及其近缘品进行判别分析，进一步构建金线莲多糖含量快速预测模型。方法 采集金线莲药材及其近缘品台湾银线兰、血叶兰样品的近红外光谱。以分类准确性为指标优选光谱预处理方法，分别对比决策树、K-近邻算法、随机森林，偏最小二乘回归判别分析、线性判别分析和支持向量机等有监督模式识别算法对样品的分类效果，优选最优算法建立定性模型。应用紫外可见分光光度法结合苯酚硫酸法测定76批金线莲多糖的含量，分别应用支持向量机、极限学习机、决策树、随机森林、主成分回归和偏最小二乘回归等6种不同化学计量学方法关联金线莲多糖含量与近红外光谱，优选定量算法；经光谱预处理和波段选择，基于连续投影算法筛选波段变量数优化并建立金线莲多糖定量模型。结果 以SNV+SG+2^ndD为光谱预处理方法结合支持向量机建立的近红外光谱判别分析模型分类准确性最高；基于径向基核函数算法结合混淆矩阵和ROC曲线评价模型的预测性能，模型性能均较优。此外，金线莲近红外原始光谱经SNV+SG+2^ndD预处理，以7 000~4 000 cm^-1为建模波段，变量数为97，应用连续投影-偏最小二乘构建的定量分析模型准确度较其他算法最高，为0.992。模型的校正集误差均方根为0.625，校正集相关系数为0.993，验证集误差均方根为0.767，验证集相关系数为0.992，预测偏差为8.467，预测集相对偏差<10%。结论 所建立的近红外支持向量机定性和连续投影-偏最小二乘定量模型准确、可靠，可鉴别金线莲真伪以及测定多糖含量，为实现金线莲质量的快速评价提供一种新的方法。     

基于深度Q网络的BECT治疗药物左乙拉西坦用药剂量推荐

目的 基于深度Q网络模型推荐伴中央颞区棘波的儿童良性癫痫患儿抗癫痫发作治疗药物左乙拉西坦的口服剂量，辅助医师制定精准的个性化用药方案。方法 收集整理2016年1月1日—2021年4月29日重庆医科大学附属儿童医院245例伴中央颞区棘波的儿童良性癫痫患儿的随访数据，利用深度强化学习技术，构建一个基于深度Q网络的儿童癫痫智能用药剂量推荐模型，并将专业医师处方与算法推荐的左乙拉西坦每日用药总剂量进行比较。结果 在推荐每日用药总剂量分布上，深度Q网络推荐的分布情况跟专业医师处方大体相似且更倾向于推荐在数据集当中分布较多并且具有统计意义的用药剂量。对基于深度Q网络剂量推荐在伴中央颞区棘波的儿童良性癫痫治疗药物左乙拉西坦的用药剂量的准确性进行了比较，每日用药总剂量分类平均准确率为89.7%，分类推荐用药平均误差为0.341 3。结论 初步验证了基于深度Q网络的用药剂量推荐模型的有效性，为该算法模型推广到更多抗癫痫发作治疗药物剂量推荐中提供参考。     

基于整合药理学平台研究葛根芩连汤治疗小儿肺炎的作用机制

目的 基于整合药理学平台探索葛根芩连汤治疗小儿肺炎的作用机制。方法 本研究通过中药整合药理学平台，通过输入葛根、黄芩、黄连、甘草四味药及肺炎获取药物和疾病靶标，然后进行蛋白质-蛋白质相互作用及网络构建，进一步进行基因功能和通路富集分析，进行葛根芩连汤治疗肺炎的"中药-成分-靶标-通路"多维网络分析。结果 药物作用于疾病的关键通路包括细胞生长和死亡、传染病、细胞凋亡、免疫系统、趋化因子信号通路、B细胞受体信号通路、T细胞受体信号通路、血小板活化、胃酸分泌等。药物潜在靶标为GCK、COX7C、ATP1A1、ADCY2。结论 葛根芩连汤治疗小儿肺炎的机制可能通过与能量代谢及消化系统相关物质唾液等相关，推测其具有潜在的关系。     

剖宫产手术预防性使用抗菌药的数据挖掘

目的 建立、比较和评价剖宫产手术抗菌药物预防使用的分类模型,为针对性干预打下基础。方法 应用数据挖掘软件PASW^® Modeler 13,建立分类模型,获得对抗菌药物预防性使用影响较大的变量(临床因素)。结果 由787例行"子宫下段剖腹产术"的病例数据获得的分类模型中,以贝叶斯网络,logistic回归和CHAID 3个模型总体较佳; 21个变量指标中,"失血量"是对该医院剖宫产手术抗菌药物预防性应用影响程度最大的因素。结论 数据挖掘技术,可以快速地建立反映剖宫产手术抗菌药物预防性使用的分类模型,为药物利用调查提供了新的分析方法。     

药物发现过程中人工智能的应用研究进展

人工智能（AI）和机器学习不仅使药物发现和开发实现了质的飞跃,而且帮助药物开发进程进入现代化。机器学习和深度学习算法已应用于药物发现各个阶段,如先导化合物的筛选、多肽合成及小分子药物的发现、最佳给药剂量的确定、类药化合物的设计和药物不良反应的预测、蛋白质间相互作用的预测、虚拟筛选效率的提高、定量构效关系（QSAR）建模和药物重新定位、理化性质和药物靶标亲和力的预测、化合物的结合预测和体内安全性分析、多靶点配体药物分子的设计以及临床试验的设计。简要综述了AI算法和传统化学相结合以提高药物发现的效率以及AI在药物发现过程中的应用研究进展,以期为AI应用于药物发现提供一定参考。     

基于系统药理学的雷公藤配伍甘草治疗类风湿性关节炎作用机制研究

目的 基于系统药理学方法研究雷公藤配伍甘草治疗类风湿性关节炎作用机制。方法 基于中药系统药理学平台（TCMSP）数据库和文献搜索,建立雷公藤配伍甘草化学成分库。运用DRAR-CPI、中医分子机制的生物信息学分析工具（BATMAN-TCM）等在线预测网站,预测成分靶标,并与疾病靶标取交集得到雷公藤配伍甘草治疗类风湿性关节炎作用靶标。通过作用靶标反向筛选潜在活性成分,并用超高效液相色谱-飞行时间质谱（UPLC-Q-TOF/MS）分析手段对活性成分进行验证。利用GeneMANIA数据库搜索间接靶标并利用蛋白互作筛选关键靶标,采用分子对接技术（SystemsDock）对潜在活性成分和关键靶标进行匹配,验证前期靶标筛选的可靠性以及反向药效团匹配的正确性。通过GO和京都基因与基因组百科全书通路分析（KEGG）生物学注释分析关键作用通路,探讨雷公藤配伍甘草治疗类风湿性关节炎作用机制。结果 共得到33个化学成分和47个潜在靶标,其中32个成分31个靶标和35条通路与药对治疗类风湿性关节炎有密切关系,主要涉及花生四烯酸代谢通路、白介素（IL）-17信号通路、核因子（NF）-κB信号通路等炎症通路以及T细胞受体信号通路、C型凝集素受体信号通路等与免疫相关的通路。结论 雷公藤配伍甘草治疗类风湿性关节炎主要通过炎症与免疫调节等多条途径得以实现。     

近红外光谱法快速鉴别不同产地的前胡药材

目的 应用近红外光谱技术结合化学计量学方法建立一种快速鉴别前胡药材产地的方法。方法 首先收集6个产地的90个前胡样本，采集各样本的近红外光谱，并划分为校正集（72个样本）和预测集（18个样本），然后利用校正集，采用化学计量学方法建立前胡药材产地的判别分析模型，最后利用预测集，对判别分析模型进行性能评价。结果 优选的判别分析模型参数如下：光谱预处理方法为多元散射校正（MSC）+Savitzky-Golay卷积二阶求导算法（SG）（窗口参数为51，拟合次数为1），光谱波段为8 400~4 200 cm^-1，判别分析模型的主成分数为18。预测集的鉴别结果表明该判别分析模型的正确识别率为100%，6个产地前胡药材之间存在明显的界限。结论 研究表明，近红外光谱法能简便、准确地实现前胡药材产地的快速鉴别，为前胡药材产地快速鉴别研究提供了理论支持和实用方法。     

HPLC-MS/MS测定新生儿血浆中氟氯西林的浓度及其与氨溴索相互作用的研究

目的 建立一种高效、简便的HPLC-MS/MS方法检测新生儿血浆中氟氯西林的浓度，研究新生儿体内氨溴索与氟氯西林是否存在相互作用。方法 生物样本采用API4000 HPLC-MS/MS进行分析，色谱柱为Ultimate XB-C₁₈(2.1 mm×100 mm，5 μm)，流动相A相为水-0.1%甲酸，B相为乙腈-0.1%甲酸。氟氯西林和内标利福平定量分析离子对分别为m/z 452.6→284.2和m/z 821.4→397.3。结果 在该分析方法下氟氯西林的线性范围为0.20~80 ng·mL^-1，定量下限为0.20 ng·mL^-1；氟氯西林的日内精密度和日间精密度均<8.23%；提取回收率为85.3%~89.2%，基质效应为89.3%~92.3%；新生儿血浆样品在室温条件下放置12 h，处理后室温条件下放置4 h，反复冻融3次及-20 ℃冷冻30 d的稳定性均良好。临床样本检测结果表明氨溴索合用可以显著提高氟氯西林血药浓度。结论 本研究所建立的HPLC-MS/MS分析方法准确度高，灵敏度好，可用于新生儿血浆中氟氯西林浓度的测定。临床研究结果表明氨溴索能显著提升氟氯西林的血药浓度，两者存在药物相互作用。     

基于药物-皮肤的相互作用探究盐酸特比萘芬体外透皮特性

目的 考察盐酸特比萘芬的体外透皮特性，探究盐酸特比萘芬与皮肤的相互作用，基于药物-皮肤相互作用阐明盐酸特比萘芬透皮特性的机制。方法 比较盐酸特比萘芬经皮渗透及其皮内滞留以及在皮肤各层的分布；利用衰减全反射红外光谱、差示扫描量热、拉曼光谱研究药物与皮肤的相互作用，并对药物与角质层角蛋白及脂质的相互作用进行计算机模拟和计算。结果 盐酸特比萘芬经皮渗透后高滞留低渗透，滞留的药物多分布于角质层。盐酸特比萘芬与角质层中脂质和角蛋白均有相互作用，该作用使药物自身难于透过皮肤，并导致较大的透过变异性。结论 盐酸特比萘芬与皮肤脂质和角蛋白的相互作用是其表现出典型的皮肤低渗透、高滞留特性的机制之一。本研究为盐酸特比萘芬体外透皮高滞留、低渗透特性提供理论依据。     

Mechanistic models enable the rational use of in vitro drug-target binding kinetics for better drug effects in patients

Introduction: Drug-target binding kinetics are major determinants of the time course of drug action for several drugs, as clearly described for the irreversible binders omeprazole and aspirin. This supports the increasing interest to incorporate newly developed high-throughput assays for drug-target binding kinetics in drug discovery. A meaningful application of in vitro drug-target binding kinetics in drug discovery requires insight into the relation between in vivo drug effect and in vitro measured drug-target binding kinetics.

Areas covered: In this review, the authors discuss both the relation between in vitro and in vivo measured binding kinetics and the relation between in vivo binding kinetics, target occupancy and effect profiles.

Expert opinion: More scientific evidence is required for the rational selection and development of drug-candidates on the basis of in vitro estimates of drug-target binding kinetics. To elucidate the value of in vitro binding kinetics measurements, it is necessary to obtain information on system-specific properties which influence the kinetics of target occupancy and drug effect. Mathematical integration of this information enables the identification of drug-specific properties which lead to optimal target occupancy and drug effect in patients.     

基于拉曼光谱法的品牌仿冒药检测方法研究

目的针对目前市面上的品牌药仿冒的制假现象,以国内多个厂家生产的头孢氨苄胶囊为工具药,建立基于支持向量机和相似度法或主成分分析的两步判别方法。方法第一步,通过模拟建立头孢氨苄胶囊品牌药及4种仿冒药的分类模型,建模交叉验证的准确率为95.63%,并对可疑样品进行预测分类,准确率为93.75%,以自制样品作为额外的预测集,假阳性率为25%。第二步,对阳性样品进行相似度计算或主成分分析,可分别将假阳性率降低至5%和0%。结果与结论两步判别法可快速、准确地实现对品牌仿冒药的检测。     

Using feature selection technique for drug-target interaction networks prediction

Elucidating the interaction relationship between target proteins and all drugs is critical for the discovery of new drug targets. However, it is a big challenge to integrate and optimize different feature information into one single "knowledge view" for drug-target interaction prediction. In this article, a feature selection method was proposed to rank the original feature sets. Then, an improved bipartite learning graph method was used to predict four types of drug-target datasets based on the optimized feature subsets. The cross-validation results demonstrate that the proposed method can provide superior performance than previous method on four classes of drug target families.     

Experimental and Computational Screening Models for Prediction of Aqueous Drug Solubility

Purpose. To devise experimental and computational models to predict aqueous drug solubility. Methods. A simple and reliable modification of the shake flask method to a small-scale format was devised, and the intrinsic solubilities of 17 structurally diverse drugs were determined. The experimental solubility data were used to investigate the accuracy of commonly used theoretical and semiexperimental models for prediction of aqueous drug solubility. Computational models for prediction of intrinsic solubility, based on lipophilicity and molecular surface areas, were developed. Results. The intrinsic solubilities ranged from 0.7 ng/mL to 6.0 mg/mL, covering a range of almost seven log₁₀ units, and the values determined with the new small-scale shake flask method agreed well with published solubility data. Solubility data computed with established theoretical models agreed poorly with the experimentally determined solubilities, but the correlations improved when experimentally determined melting points were included in the models. A new, fast computational model based on lipophilicity and partitioned molecular surface areas, which predicted intrinsic drug solubility with a good accuracy (R ²of 0.91 and RMSE_tr of 0.61) was devised. Conclusions. A small-scale shake flask method for determination of intrinsic drug solubility was developed, and a promising alternative computational model for the theoretical prediction of aqueous drug solubility was proposed.     

基于系统药理学探索裸花紫珠有效成分的作用机制

目的 通过系统药理学方法探索裸花紫珠有效成分的药理作用机制。方法 运用文献搜索和TCMSP数据库,建立裸花紫珠活性分子数据库,通过对数据库中的120个分子从口服利用度、类药性分析等几个方面进行筛选,预测出有潜在活性的分子,并通过王永华教授团队研发的SYSDT以及WES技术进行靶标预测,进而通过药物-靶标预测模型Sys TD、Drug Bank数据库等构建靶标-疾病-通路网络图,并借助基因本位论（GO）分析裸花紫珠有效成分参与的生物学过程。结果 裸花紫珠中黄酮类化合物如木犀草素及其衍生物、甲基鼠李素和芹菜素等作用于较多的靶点,在裸花紫珠的药效中起着关键作用;其对炎症、癌症、心血管疾病、神经系统以及免疫系统均有影响,其中ABCB1、NOS3、MAPK14、PPARG、GSK3β、PTGS2等同时靶向治疗多种类型疾病和多种生物学通路。GO分析表明了裸花紫珠对炎症反应、凝血、血管生成、钙离子信号传导、血压等具有调节作用。结论 裸花紫珠黄酮类化合物在的药效中起关键作用,通过ABCB1、NOS3、MAPK14等靶点对炎症反应、凝血、血管生成、钙离子信号传导、血压等发挥调节作用。     

TargetHunter: An In Silico Target Identification Tool for Predicting Therapeutic Potential of Small Organic Molecules Based on Chemogenomic Database

Target identification of the known bioactive compounds and novel synthetic analogs is a very important research field in medicinal chemistry, biochemistry, and pharmacology. It is also a challenging and costly step towards chemical biology and phenotypic screening. In silico identification of potential biological targets for chemical compounds offers an alternative avenue for the exploration of ligand–target interactions and biochemical mechanisms, as well as for investigation of drug repurposing. Computational target fishing mines biologically annotated chemical databases and then maps compound structures into chemogenomical space in order to predict the biological targets. We summarize the recent advances and applications in computational target fishing, such as chemical similarity searching, data mining/machine learning, panel docking, and the bioactivity spectral analysis for target identification. We then described in detail a new web-based target prediction tool, TargetHunter (http://www.cbligand.org/TargetHunter). This web portal implements a novel in silico target prediction algorithm, the Targets Associated with its MOst SImilar Counterparts, by exploring the largest chemogenomical databases, ChEMBL. Prediction accuracy reached 91.1% from the top 3 guesses on a subset of high-potency compounds from the ChEMBL database, which outperformed a published algorithm, multiple-category models. TargetHunter also features an embedded geography tool, BioassayGeoMap, developed to allow the user easily to search for potential collaborators that can experimentally validate the predicted biological target(s) or off target(s). TargetHunter therefore provides a promising alternative to bridge the knowledge gap between biology and chemistry, and significantly boost the productivity of chemogenomics researchers for in silico drug design and discovery.     

Development of a decision tree to classify the most accurate tissue-specific tissue to plasma partition coefficient algorithm for a given compound

Physiologically based pharmacokinetic (PBPK) modeling is a tool used in drug discovery and human health risk assessment. PBPK models are mathematical representations of the anatomy, physiology and biochemistry of an organism and are used to predict a drug’s pharmacokinetics in various situations. Tissue to plasma partition coefficients (Kp), key PBPK model parameters, define the steady-state concentration differential between tissue and plasma and are used to predict the volume of distribution. The experimental determination of these parameters once limited the development of PBPK models; however, in silico prediction methods were introduced to overcome this issue. The developed algorithms vary in input parameters and prediction accuracy, and none are considered standard, warranting further research. In this study, a novel decision-tree-based Kp prediction method was developed using six previously published algorithms. The aim of the developed classifier was to identify the most accurate tissue-specific Kp prediction algorithm for a new drug. A dataset consisting of 122 drugs was used to train the classifier and identify the most accurate Kp prediction algorithm for a certain physicochemical space. Three versions of tissue-specific classifiers were developed and were dependent on the necessary inputs. The use of the classifier resulted in a better prediction accuracy than that of any single Kp prediction algorithm for all tissues, the current mode of use in PBPK model building. Because built-in estimation equations for those input parameters are not necessarily available, this Kp prediction tool will provide Kp prediction when only limited input parameters are available. The presented innovative method will improve tissue distribution prediction accuracy, thus enhancing the confidence in PBPK modeling outputs.