Prediction of gene functions by use of molecular evolutionary approach

本研究的主要目的是利用分子进化的方法，从同源基因群的核苷酸序列中提取功能结构域。为了达到这一目的，我们首先发展了基于最小描述长度原理的构建系统发育树的新方法。利用这种新的方法，我们以丝氨酸蛋白酶基因的几个功能结构域为例进行了鉴定。由于这种方法非常成功，我们将我们的方法扩展到所有同源基因组。然后，我们构建了所谓的序列基序数据库，其中序列基序表示特定功能的序列的特征。此外，通过进行聚合酶链式反应实验，我们产生了数据可获得性有限的基因的核苷酸序列。为了建立本实验室的测序技术，我们利用鱼类线粒体DNA进行了聚合酶链式反应实验。这些研究结果对于从核苷酸序列数据中识别和预测基因功能具有重要意义。

The main purpose of the present research project is to extract functional domains from the nucleotide sequences of homologous gene groups by use of molecular evolutionary approach. To attain this purpose, we first developed the the new method for constructing phylogenetic trees based upon the minimum description length principle. By using this new method, we identified several functional domains of serine protease genes as an example. Because this method was so successful, weextended our approach to all homologous gene groups. Then, we constructed the 'so-called' sequence motif database, where sequence motif represents characteristics of the sequence specific to a particular function. Moreover, we produced nucleotide sequences of genes whose data availability is sos limited by conducting the PCR experiments. In order to establish the sequence techniques in our laboratory, we practiced the PCR experiment by using fish mitochondrial DNAs. These results obtained in this research project is significant for identifying and predicting the gene functions from the nucleotide sepuence data.