Data-Farming: Systemic Tree-like Infrastructure Development of the Data- and Knowledge Bases on functional DNA and RNA sites

摘要

At molecular level, life-processes are realized through interactions between DNA, RNA, and proteins. The interacting fragments are denoted as "functional sites"; the site clusters - as "regulatory regions"; the protein coding DNA(RNA) regions as "genes". For DNA(RNA) sequences, Genome Annotation is a field of life-science predicting of functional sites, regulatory regions, and genes, which should be verified experimentally. So, the applets predicting the positioning of above described structural units by similarity to the already known patterns are needed. All the experimental evidence on DNA(RNA) sequences, sites, regulatory regions, genes is distributed within over 400 databases. To apply these data for Genome Annotation, the complementary data portions should be retrieved from the databases and treated by the "Knowledge Discovery" computations in order to develop the applets suitable for the sequence analysis. Since a number of various "Knowledge Discovery" computations contain some common steps, we have stored the intermediate results obtained at these steps within an intermediate database, called the "data-filial" database. This yielded the data-filial tree, called "data-farming", growing from the databases to the applets for sequence analysis. Our data-farming infrastructure for (ⅰ) functional site activity prediction, (ⅱ) gene regulatory region recognition, and (ⅲ) "High/Low"- gene expression estimate, is described and discussed here, URL=.