Knowledge Discovery for Molecular Biology: Algorithms and Experiments

这是三年持续奖励的第一年资助。分子生物学中信息量的迅速增长对自动化分析提出了不断增长的需求。为了有效地从这些数据库中提取知识，计算机科学家必须与其他领域的研究人员一起开发这种分析的工具，因为这些新数据库的复杂性使传统的人工分析技术过时了。该项目将开发新的算法，在必要时整合旧算法，并将这些算法应用于从科学数据库中提取知识的问题，特别是那些由分子生物学研究人员组装的知识。这些算法的优点和缺点将通过仔细的实验来探索。所开发的算法将采用分类、聚类和搜索技术，这些技术可以推广到其他问题领域

This is the first year funding of a three-year continuing award. The rapidly growing stores of information in molecular biology present an ever-increasing demand for automated analysis. In order to extract knowledge from these databases efficiently, computer scientists must join with researchers in other fields to develop tools for this analysis, since the complexity of these new databases antiquates traditional manual analysis techniques. This project will develop new algorithms, integrating old ones where necessary, and apply these algorithms to the problem of extracting knowledge from scientific databases, especially those being assembled by researchers in molecular biology. The strengths and weaknesses of these algorithms will be explored through careful experimentation. The algorithms developed will employ techniques of classification, clustering, and search which may be generalizable to other problem domains.//