几种改进的CoMFA方法比较研究血小板活化因子拮抗剂

由于传统的比较分子场分析(CoMFA)方法本身存在一些缺陷，使得分子的叠合 规则以及叠合分子的空间取向和空间位置等因素对q~2的影响很大，因此相继提出 了几种改进的CoMFA方法。为了优化CoMFA结果，应用传统的CoMFA方法和交叉验证 的R~2引导的区域选择法(q~2-GRS)、全取向搜索法（AOS）、全空间搜索法（APS） 以及比较分子相似性指数(CoMSIA)等四种改进的CoMFA方法，对18个pinusolide类 衍生物这类新发现的血小板活化因子（PAF）拮抗剂进行了比较研究。结果表明四 种改进的CoMFA方法得到的q~2值均比传统CoMFA的高。q~2-GRS方法得到的q~2值有 所提高，但综合结果并不理想，AOS与APS得到的q~2较为理想，而在CoMSIA中， q~2几乎不受空间取向或空间位置的影响。同时我们引人基于样本的偏最小二乘法 （SAMPLS）取代原AOS／APS程序中的传统PLS进行统计分析，明显提高了其运行速 度。最后，根据q~2最高的CoMFA模型和CoMSIA模型设计了几个预测活性更高的 pinusolide类似物。     

N－亚硝基化合物的结构／致癌活性的三维构效关系研究

用比较力场分析研究了Ｎ－亚硝基化合物的结构与致癌活性的关系，考察了网络结构和探针原子对结果的影响．结果表明，立体效应和静电作用场是描述其致癌活性和进行结构性能关系研究的最重要的结构参数。     

环氧合酶－2抑制剂的三维定量构效关系研究

建立三环系COX－1和COX－2抑制剂结构与活性的三维定量构效关系模型，为设 计新型的具有选择性的COX－2抑制剂提供线索。通过与酶的对接并优化，确定化合 物在受体结合腔中的构象，利用比较分子力场分析方法建立三维定量构效关系模型 。模型1R_(cv)~2=0.685，最佳主成分数为6，传统相关系数为R~2=0.988, F-726. 2，标准偏差S = 0.080；模型2 R_(cv)~2 = 0.573，最佳主成分数为6，传统相关 系数为R~2=0.996, F = 1147.6，标准偏差S = 0.034。所得的模型不仅解释了化合 物的构效关系，而且对预测集中的化合物有很好的预测能力；比较不同模型的系数 相关图，分析了结构与活性、结构与选择性的关系，得到的结果可以指导新化合物 的设计与合成。     

Comparative molecular field analysis (CoMFA) for phosphodiesterase (PDE) IV inhibitors

A comparative molecular field analysis (CoMFA) for phosphodiesterase (PDE) IV inhibitors has been performed to correlate their chemical structures with their observed biological activity. In this study, CoMFA model based on docking mode for active site of PDE IV can describe the relative change in magnitude of the steric and electrostatic fields as a function of the compounds. Pyridine N-oxide and pyridine group of each compound are aligned toward the metal ion in S2-sub pocket of PDE IV. The study provided a statistically valid model with good correlation and predictive power, and consequently we identified some key features that may be used to design new derivatives.     

喹诺酮类N1位定量构效关系研究

氟隆诺酮类化合物是抗菌药中的重要一类，由于它的许多突出特点，已引起越来越多的研究者的关注山．这类化合物骨架见图1．我们对该类化合物进行过一些研究［’‘l自发现诺氟沙星＊以后，几位取代基在寻找高效、广谱峻诺酮类化合物方面起重要作用，但在发现环丙基（1983年）、特丁基（1989年）以后，一直未发现新的有意义的取代基·在NI位定性与定星构效关系的研究中有二种观点：（1）古贺弘等k’］提出的由立体效应决定的观点；（z）Domagala［71、Chu问等主张的立体与电子效应共同起作用的观点·3D－QSAR研究中的比较分子力场分析（Co…     

褪黑激素受体拮抗剂的三维定量构效关系研究

采用两种分子场分析方法即比较分子场分析法（CoMFA）和比较分子相似因子分析法（CoMSIA）进行了37个褪黑激素受体拮抗剂的构效关系研究.计算结果表明，两种方法得到的构效关系模型都具有较好的预测能力.在计算中，还考察了不同格点距离和电荷计算方法对构效关系模型的影响.通过分析分子场等值面图在空间的分布，可以观察到叠合分子周围分子场特征对化合物活性的影响，为设计新的褪黑激素拮抗剂提供了一些理论依据.     

An improved nicotinic pharmacophore and a stereoselective CoMFA-model for nicotinic agonists acting at the central nicotinic acetylcholine receptors labelled by [3H]-N-methylcarbamylcholine

A study of a series of compounds with agonistic effect at the alpha4beta2 nicotinic acetylcholine receptors resulted in an improved pharmacophore model as well as a CoMFA model. The pharmacophore was composed of three pharmacophoric elements: (1) a site point (a) corresponding to a protonated nitrogen atom, (2) a site point (b) corresponding to an electronegative atom capable of forming a hydrogen bond, and (3) the centre of a heteroaromatic ring or a C=O bond (c). The pharmacophoric elements were related by the following parameters: (a-b) 7.3-8.0 A, (a-c) 6.5-7.4 A, and the angle between the two distance vectors (delta bac) 30.4-35.8 degrees. In addition to this, a stereoselective CoMFA model was developed, which showed good predictability even for compound classes not present in the training set.     

3 D-QSAR Analysis of Agonists of nAChRs： Epibatidine Analogues

A 3 D-QSAR about nAChRs agonists epibatidine analogues was performed using theCoMFA and CoMSIA. The correlation coefficients were R2cv = 0.546, R2cv = 0.907 in CoMFA andR2cv = 0.655, R2,~ = 0.962 in CoMSIA of the final model. The prediction using the final models tothe test set was r2 = 0.675 in CoMFA and r2 = 0.462 in CoMSIA. This model will be useful in thedesign of novel compounds with high affinity.     

GABAA五种亚型受体与BZ配基的3D-QSAR研究

GABAA受体是中枢神经系统内重要的抑制性受体,有广泛的神经生理活性.由于镇静/抗惊厥药物在临床上的广泛应用,使得其中苯并二氮杂作用位点尤为重要.我们用比较分子场法(CoMFA)对一系列咪唑苯并二氮杂类化合物（BZ）与五种重组受体亚型的亲和力进行了结构活性关系研究,得到的一组模型都有较高的交叉验证系数.并在此基础上,建立了非交叉验证的一组PLS模型.用该组模型对随机选择的6个化合物组成的测试集进行了预测,都得到了相当满意的结果,表明所建立的一组模型具有良好的预测能力.本研究对于设计高亲和力的BZ受体的配基和研究GABAA受体的模型有指导意义.     

Effects of variable selection on CoMFA coefficient contour maps in a set of triazines inhibiting DHFR

Summary An example of a CoMFA study is described with the aim to discuss one of the major problems of this 3D QSAR method: lack of variable selection. It is shown that the use of nonrelevant energy parameters might produce CoMFA contour maps which poorly reflect the actual nature of the binding site and are in part statistical artefacts. The data set employed in our analysis comparises triazine inhibitors of dihydrofolate reductase (DHFR), isolated from chicken liver, which have already been the object of a QSAR study by other authors. Since three-dimensional structures of triazine-DHFR complexes are known, it was possible not only to reduce ambiguities in the superimposition of the ligands, but also to compare the resulting CoMFA contour maps with the enzyme active site.Supplementary material available: The Cartesian coordinates and the atomic charges of the PM3-optimized structures used in the CoMFA study are available as MOL2 files upon request.To whose memory this paper is dedicated.