化合物结构特征与其基因表达信息的互补性研究

Complementary Study of Structure Features and Gene Profile Features for Chemical Compounds

化合物分子指纹在化合物相似性比较、生物芯片技术在分子标记识别中都有着重要的应用,因而受到广泛的重视.但是由于这两个研究对象依传统的观点,分属于化学信息学和生物信息学两个不同的研究领域,因此一直未能在同一研究对象中同时得到应用.近年来,小分子刺激的生物芯片数据的大量出现,使得联合化合物的结构特征及其基因表达谱信息共同的用于药物筛选成为可能.为此,根据本质和外延互补的哲学思想,提出了综合运用这两种数据来共同描述化合物的新方法.通过(1)只利用化合物分子指纹,(2)只利用化合物基因表达谱信息和(3)前面两种方法互补等三种方法在化合物相似性比较的研究中,发现本文提出的互补方法比前两种方法的查询结果更准确:(1)使结构和功能两方面均相似的化合物之间的相似性更高,(2)并且能有效使那些仅在某一方面表现非常特殊的化合物能够暴露出来,而得到有效地排除.本研究给化学信息者和生物信息者重新审视以往的数据,联合致力于药物的筛选提供了线索.化合物分子指纹(本质)和基因表达谱(外延)的互补原理则为药物的筛选提供更有效的依据,并最终加快药物的研发进程.

The application of molecular fingerprints has a long history in computational medicinal chemistry,on the other side,gene microarray is playing an important role in bio-marker identification.Traditionally,these two technologies belong to different areas,i.e.,chemoinformatics and bioinformatics,respectively.And they have never been jointly applied in one study yet.Recently,the appearance of small-molecular microarray data makes it possible to jointly use the compound structure information as well as gene-expression profile as a comprehensive description for drug study.Therefore,a novel approach is reported to describe compounds by integrating data from both of the two types,on the basis of the philosophy of the mutual complement of ＂essence＂ and ＂extension＂.Comparing of the similarity searching results derived from the compound description with only the structure fingerprint or the gene-expression profile,our integrated methodology indicated much better description ability.As a conclusion,our complementary description of chemical compound from an integration view of structure fingerprint and gene expression file will shed new light on the drug study in both area of bioinformatics and chemoinformatics.It is expected to provide more effective drug screening model with its great description ability and accelerate the process of drug design eventually.