商业化合物和天然产物的化学信息学分析

发布时间：2018-05-14 14:18

本文选题：商业化合物库 + 陆生天然产物和海洋天然产物　；参考：《华中农业大学》2017年硕士论文

【摘要】：随着计算水平(规模和精度等)的飞速提高,计算机辅助药物设计(computer-aided drug design,CADD)在药物设计中扮演越来越重要的地位。其从基于配体的药物设计(ligand-based drug design,LBDD)和基于结构/受体的药物设计(structure/receptor-based drug design,SBDD)两种角度利用虚拟筛选(virtual screening,VS)大大缩短了药物研发时间,并显著降低了高通量筛选(high-throughput screening,HTS)等所需的花费。而各种供应商提供的大型可购买小分子筛选库是虚拟筛选不可或缺的来源,其中含有的小分子即本研究所称的商业化合物。为特定的VS流程选择最佳筛选库,对于提高成功率和避免在后期实验阶段浪费资源是非常重要的。因此分析不同筛选库中分子的结构特征和多样性是很有必要的。同时,这又是基于海洋或陆地来源的天然产物的药物发现的新黄金时期。各种挖掘天然产物价值的研究层出不穷,大大推动了天然产物在实际药物设计过程中的使用。而海洋天然产物(marine natural products,MNPs)和陆地天然产物(terrestrial natural products,TNPs)之间的结构差异是一个十分有趣而关键的问题。所以,本论文对一些代表性商业数据库和许多天然产物库进行了基于结构的化学信息学分析。在本研究中,首先,我们对11种可购买筛选库和中药化合物数据库(Traditional Chinese Medicine Compound database,TCMCD)的结构特征和骨架多样性进行了分析比较。其骨架由Murcko骨架和Level 1 Scaffold Tree表征,其多样性特征通过骨架计数和累积骨架频率曲线来刻画,并通过Tree Maps和SAR Maps进行可视化。分析表明,基于具有相似分子量分布的标准化子集,Chembridge,ChemicalBlock,Mucle,TCMCD和VitasM在结构上比其他数据库更加多样化。与所有可购买的筛选库相比,TCMCD具有最高的结构复杂性,但更保守的分子骨架。此外,我们发现一些代表性骨架是针对不同药物靶标的药物候选物的重要组成部分,例如激酶和G蛋白偶联受体(G protein-coupled receptors,GPCRs),因此在筛选库中含有这些药骨架的分子可能是相关靶标潜在的抑制剂。这启发我们哪些可购买的化合物库在相应的VS中可能产生更好的结果其次,我们分析了两种天然产物(MNPs和TNPs)的重要物理化学性质,结构特征和类药性,并从进化角度对其差异进行了讨论。结果显示,MNPs具有更低的溶解度。从独特的片段和骨架的角度看,MNPs具有更长的长链且通常形成大环,特别是8至10元环。在MNPs中发现了更多的氮原子,卤素(特别是溴原子)和更少的氧原子,这表明MNPs可以通过比TNPs更多样化的生物合成途径合成,并且早期地球上的氧浓度影响代谢进化。而TNPs比MNPs具有更多的五或六元环,表明它们在结构上更稳定。与陆地化合物相比,海洋化合物中高达76.83%的Murcko骨架是独特的。此外,综合来看,尽管两种天然产物的类药性评估很接近,MNPs(78.06%)比TNPs(76.61%)有稍高的类药性,这可以通过进化观点来解释:古老的物种(如海洋中的许多物种)由于生物体和分子之间的共进化作用而能够提供更多的靶标化合物(hits)。但考虑到已有许多TNPs被开发,以及MNPs的巨大潜力,我们相信MNPs可能给从事药物发现的研究人员带来更多的希望。同样的,本文中提到的TNPs的特征也会在今后的药物设计中起到一定作用。
[Abstract]:With the rapid improvement of computing level (scale and precision), computer-aided drug design (CADD) plays a more and more important role in drug design. It is from ligand based drug design (ligand-based drug design, LBDD) and structural / receptor based drug design (structure/receptor-based drug DESI) GN, SBDD) two angles using virtual screening (VS) greatly shortens the time of drug development and significantly reduces the cost required for high throughput screening (high-throughput screening, HTS). And the large purchase small molecule screening library provided by various suppliers is an indispensable source of virtual screening, which contains small molecules. The selection of the best screening library for a specific VS process is important for improving the success rate and avoiding the waste of resources at the later stage of the experiment. Therefore, it is necessary to analyze the structural characteristics and diversity of the molecules in different screening libraries, and it is also based on the natural production of marine or terrestrial sources. The new gold period of the drug discovery. Various studies on the value of natural products are emerging, greatly promoting the use of natural products in the actual drug design process. The structural differences between the marine natural products (marine natural products, MNPs) and the natural products of the land (terrestrial natural products, TNPs) are very important. Interesting and critical issues. So, this paper carries out structural based chemical informatics analysis on some representative commercial databases and many natural product libraries. In this study, first, we have the structural features of the 11 kinds of Traditional Chinese Medicine Compound database (TCMCD) database and the traditional Chinese medicine compound database (Medicine Compound database, TCMCD). Skeleton diversity was analyzed and compared. The skeleton was characterized by Murcko skeleton and Level 1 Scaffold Tree. The diversity of the skeleton was characterized by the skeleton count and the cumulative skeleton frequency curve, and was visualized through Tree Maps and SAR Maps. The analysis showed that the standard subset, Chembridge, ChemicalBlock based on the similar molecular weight distribution, was analyzed. Mucle, TCMCD and VitasM are structurally more diverse than other databases. Compared with all the available screening libraries, TCMCD has the highest structural complexity, but a more conservative molecular skeleton. In addition, we find that some representative skeletons are important components of drug candidates for different drug targets, such as kinases and G proteins. G protein-coupled receptors (GPCRs), so the molecules containing the skeleton of these drugs in the screening library may be potential inhibitors of the related target. This enlightens us which can produce better results in the corresponding VS. We analyze the important physicochemical of two natural products (MNPs and TNPs). The results show that MNPs has a lower solubility. From the point of view of unique fragments and skeletons, MNPs has longer long chains and usually forms large rings, especially 8 to 10 yuan rings. More nitrogen atoms, halogens (especially bromine atoms), are found in MNPs. Less oxygen atoms, which indicates that MNPs can be synthesized by more than TNPs biosynthetic pathway, and the oxygen concentration on the early Earth affects metabolic evolution. TNPs has more five or six membered rings than MNPs, indicating that they are more structurally stable. Compared with terrestrial compounds, 76.83% of the Murcko skeleton in marine compounds is the only one. In addition, in addition, in spite of the close proximity of the two natural products, MNPs (78.06%) is slightly higher than TNPs (76.61%), which can be explained by the evolutionary perspective: the ancient species (such as many species in the ocean) can provide more target compounds due to the co evolution between organisms and molecules (HI TS). But considering the development of many TNPs and the great potential of MNPs, we believe that MNPs may bring more hope to researchers engaged in drug discovery. Similarly, the features of TNPs mentioned in this article will also play a role in future drug design.

【学位授予单位】：华中农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TQ460.1;O629

【相似文献】