基于RP-HPLC-DAD指纹图谱结合化学计量学评价不同品种木豆叶质量及主要成分含量测定

目的 基于RP-HPLC-DAD指纹图谱结合化学计量学方法,评价不同品种木豆Cajanus cajan叶质量特征的共有性和差异性,为木豆叶的质量评价提供科学依据。方法 通过RP-HPLC-DAD建立13个不同品种木豆叶的RP-HPLC-DAD指纹图谱,进行相似度分析（similarity analysis,SA）,结合聚类分析（hierarchical cluster analysis,HCA）、主成分分析（principalcomponent analysis,PCA）和最小二乘判别分析（partial least squares discriminant analysis,PLS-DA）对木豆叶质量进行综合评价,并对样品中6种主要成分进行含量测定。结果 木豆叶RP-HPLC-DAD指纹图谱共标定13个共有峰,13个品种木豆叶的相似度均在0.90以上,通过HCA可将这13个品种的木豆叶分为5类。PCA与HCA结果基本一致,经过PCA分析发现S8的综合得分最高,质量最好,其次是S9和S12。PLS-DA与PCA结果基本一致,不同品种木豆叶的化学成分差异主要由7、10、6、9、11号5个色谱峰引起的。S8、S9和S12的含量处于前3,与PCA分析的综合得分一致。结论 指纹图谱结合HCA、PCA和PLS-DA可以全面的评价木豆叶质量,为木豆叶的质量控制提供全面的参考。     

An NMR based panorama of the heterogeneous biology of acute respiratory distress syndrome (ARDS) from the standpoint of metabolic biomarkers

Acute respiratory distress syndrome (ARDS), manifested by intricate etiology and pathophysiology, demands careful clinical surveillance due to its high mortality and imminent life support measures. NMR based metabolomics provides an approach for ARDS which culminates from a wide spectrum of illness thereby confounding early manifestation and prognosis predictors. ¹H NMR with its manifold applications in critical disease settings can unravel the biomarker of ARDS thus holding potent implications by providing surrogate endpoints of clinical utility. NMR metabolomics which is the current apogee platform of omics trilogy is contributing towards the possible panacea of ARDS by subsequent validation of biomarker credential on larger datasets. In the present review, the physiological derangements that jeopardize the whole metabolic functioning in ARDS are exploited and the biomarkers involved in progression are addressed and substantiated. The following sections of the review also outline the clinical spectrum of ARDS from the standpoint of NMR based metabolomics which is an emerging element of systems biology. ARDS is the main premise of intensivists textbook, which has been thoroughly reviewed along with its incidence, progressive stages of severity, new proposed diagnostic definition, and the preventive measures and the current pitfalls of clinical management. The advent of new therapies, the need for biomarkers, the methodology and the contemporary promising approaches needed to improve survival and address heterogeneity have also been evaluated. The review has been stepwise illustrated with potent biometrics employed to selectively pool out differential metabolites as diagnostic markers and outcome predictors. The following sections have been drafted with an objective to better understand ARDS mechanisms with predictive and precise biomarkers detected so far on the basis of underlying physiological parameters having close proximity to diseased phenotype. The aim of this review is to stimulate interest in conducting more studies to help resolve the complex heterogeneity of ARDS with biomarkers of clinical utility and relevance.     

基于HPLC指纹图谱和化学计量学的不同产地雪菊质量评价

目的 建立雪菊HPLC指纹图谱，结合化学计量学，测定4种成分含量，评价不同产地雪菊质量。方法 采用依利特SinoChrom ODS-BP C₁₈色谱柱（4.6 mm×250 mm，5 μm），以乙腈-0.1%磷酸溶液为流动相，梯度洗脱；流速为0.6 mL·min^-1；检测波长为295 nm；柱温为30℃；建立不同产地雪菊指纹图谱，进行相似度评价、聚类分析、主成分分析、偏最小二乘法-判别分析，结合变量重要性投影值筛选指认主要差异成分，进行含量测定。结果 12批雪菊指纹图谱含21个共有峰，所有样品与对照图谱相似度均≥0.942；经聚类分析和主成分分析，12批雪菊聚为3类；偏最小二乘法-判别分析结果与聚类分析一致，变量重要性投影值筛选的4个主要差异成分含量分别为马里苷40.05~61.25 mg·g^–1，黄诺马苷7.44~19.82 mg·g^–1，芦丁0.85~2.03 mg·g^–1，绿原酸2.56~9.73 mg·g^–1。结论 建立的指纹图谱分析与含量测定方法可靠、重复性好，为雪菊的整体质量评价提供了参考。     

Quality Evaluation and Regional Analysis of Psoraleae Fructus by HPLC-DAD-MS/MS plus Chemometrics

Objective To evaluate the quality of Psoralea corylifolia collected from 12 provinces of China. Methods An HPLC-DAD-MS/MS method was used to identify, determine, and estimate 14 representative bioactive compounds in P. corylifolia. Then on the basis of the content data, the chemometrics method was used to differentiate 20 samples from different regions. Results The quality of P. corylifolia from 12 different provinces of China was evaluated by this method. Though the samples showed similar profiles, content of the detected markers varied significantly in different regions and batches. According to the results of the hierarchical cluster analysis and principle component analysis, it can be concluded that the samples from different origins could be clustered reasonably into two groups, as well as successfully distinguished. Conclusion A simple and reliable new method which used HPLC-DAD-MS/MS and chemometrics has been developed to characterize, classify, and control the quality of P. corylifolia.     

基于一测多评(QAMS)法联合化学计量学的参贝止咳颗粒综合质量评价

目的 建立参贝止咳颗粒中重楼皂苷Ⅶ、Ⅵ、Ⅰ及白花前胡甲素、白花前胡乙素、白花前胡素E、法卡林二醇、人参炔醇、柚皮芸香苷、橙皮苷、川陈皮素、橘红素的高效液相色谱(HPLC)一测多评(quantitative analysis of multi-components by single-marker,QAMS)方法,并结合...     

Optimal classification of long echo time in vivo magnetic resonance spectra in the detection of recurrent brain tumors

We describe the optimal high-level postprocessing of single-voxel (1)H magnetic resonance spectra and assess the benefits and limitations of automated methods as diagnostic aids in the detection of recurrent brain tumor. In a previous clinical study, 90 long-echo-time single-voxel spectra were obtained from 52 patients and classified during follow-up (30/28/32 normal/non-progressive tumor/tumor). Based on these data, a large number of evaluation strategies, including both standard resonance line quantification and algorithms from pattern recognition and machine learning, were compared in a quantitative evaluation. Results from linear and non-linear feature extraction, including ICA, PCA and wavelet transformations, and also the data from resonance line quantification were combined systematically with different classifiers such as LDA, chemometric methods (PLS, PCR), support vector machines and ensemble methods. Classification accuracy was assessed using a leave-one-out cross-validation scheme and the area under the curve (AUC) of the receiver operator characteristic (ROC). A regularized linear regression on spectra with binned channels reached 91% classification accuracy compared with 83% from quantification. Interpreting the loadings of these regressions, we find that lipid and lactate signals are too unreliable to be used in a simple machine rule. Choline and NAA are the main source of relevant information. Overall, we find that fully automated pattern recognition algorithms perform as well as, or slightly better than, a manually controlled and optimized resonance line quantification.     

HPLC指纹图谱结合化学计量学与抗氧化能力评价不同产地槐米的品质

目的以连续2年收集于不同产地的槐米样品为材料,建立槐米HPLC指纹图谱,并采用HPLC法同时测定5种黄酮类组分含量,结合总黄酮含量及抗氧化能力评价不同产地槐米的品质。方法采用HPLC法和分光光度法对已收集54份槐米样品的芦丁、槲皮素、染料木素、山柰酚、异鼠李素和总黄酮组分含量及DPPH、羟自由基、超氧阴离子清除能力和脂质过氧化抑制能力等指标进行测定,运用相似度、聚类分析、主成分分析方法综合分析不同产地槐米的品质。结果建立了HPLC指纹图谱,样品组分分离良好,方法学考察均符合要求,共有峰28个,确定了峰面积之和占共有峰总峰面积80%以上的5个组分;以总黄酮和5个黄酮类组分的含量以及4个抗氧化能力指标为依据,将54份样品进行聚类分析和主成分分析,两者结果一致,不同产地的样品没有表现出显著的地域特性。结论 HPLC结合化学计量学及抗氧化能力可准确、综合地对槐米进行品质评价,通过分析表明不同生产年度同一产地间的样品品质有差异,同一生产年度不同产地间样品品质有差异,说明槐米品质受气候的影响;还表明同一生产年度同一产地的不同样品品质也有差异,充分说明了槐米的品质也受关键生产技术环节的影响。     

试论主成分分析在中药研究中的应用

主成分分析在中药研究中的应用存在一定的问题。本文以具体实例讲解了主成分分析的使用方法,并总结了主成分分析的注意事项。     

延胡索药材HPLC指纹图谱的化学模式识别研究

目的：以21种不同产地的延胡索中药色谱指纹图谱为基础，采用化学计量学方法对其进行化学模式识别，并对不同的化学模式识别方法进行比较。方法：将指纹图谱信息进行预处理，经过峰对齐及标准化后，用MATLAB数据处理软件进行模式识别分析。结果：对21种样品进行了分类和鉴别，采用此方法所得结果与实际相符合。结论：本文方法结果可靠，可用于不同产地的中药延胡索的分类和鉴别。     

紫外可见光谱结合化学计量法快速检测双黄连口服液大类成分

目的 基于紫外可见光谱和化学计量学方法建立1种快速检测双黄连口服液质量的方法。方法 采集双黄连口服液紫外可见光谱数据,并使用主成分分析筛选剔除异常样本数据。采用Kennard-Stone算法将所有样本按照7∶3的比例划分为训练集和测试集。采用一阶求导、标准正态变量变换对数据进行预处理,然后结合竞争性自适应重加权采样法提取特征波长,最后将支持向量回归(support vector regression,SVR)、最小二乘支持向量回归(least squares of support vector regression,LS-SVR)、前馈型反向传播(back propagation,BP)神经网络3种方法用于可溶性固形物(soluble solids content,SSC)、总黄酮(total flavones,TF)的定量分析模型的建立。结果 3种模型的决定系数R²均≥0.816 8,均方根误差RMSE均≤4.378 2,均获得了较好的预测效果。对测试集SSC及TF预测结果进行对比发现,与BP神经网络、LS-SVR相比,SVR模型获得了最大R²以及最小RMSE。SVR-SSC模型的R²为0.999 8,RMSE为0.260 3,SVR-TF模型的R²为0.998 3,RMSE为0.543 3。结论 紫外可见光谱结合SVR可以提供1种双黄连口服液质量的高精度快速现场检测方法。