REPRESENTATION AND RETRIEVAL OF PHYSICAL MAPPING DATA

我们建议开发一个用于存储、检索和图形的数据库 展示几个实验室正在制作的物理测绘数据 耶鲁大学人类遗传学和生物系。在不久的将来, 描述人类基因组的大量物理图谱数据将被 在耶鲁大学和其他地方生产，并将需要存储在有组织的 时尚，以便它可以很容易地检索和显示。许多不同的类型 的数据正在产生，这些数据描述了不同的基因组 细节层次。这些信息通常是不确定和不完整的， 并且在一个级别的数据可以与(加强、补充或 矛盾)另一个层面的数据。检索此数据的例程 应智能地适应不确定性、不完全性和 数据的相互依赖关系。 物理映射数据的表示形式。肯尼斯博士的实验室 基德，大卫·沃德，谢尔曼·魏斯曼，弗兰克·拉德尔，斯蒂芬·里德斯，以及 艾伦·贝尔目前正在制作大量的物理地图 数据，包括大比例尺限制图、重叠的宇宙集合、 酵母人工染色体作图和原位杂交数据。 我们将为这些数据开发和改进适当的表示法。我们 将特别侧重于如何最好地表示固有的不确定性 和数据中的相互依赖关系，以及基于 数据。 数据的智能检索和简明显示。我们将发展 例行公事，融合了几种类型的知识，使智能 数据的检索。我们还将开发例程来显示数据 以不同的细节级别以图形方式显示，包括 基因图谱。尽管数据库本身中的数据会很复杂，而且 相互关联的显示例程将被设计为呈现数据 显然，在适当的细节水平上。例如，一张地图 染色体或大或小的染色体区域将是简洁的和 很容易理解。

We proposed to develop a database for storage, retrieval, and graphical display of physical mapping data being produced by several laboratories at the Yale Departments of Human Genetics and Biology. In the near future, large amounts of physical mapping data describing the human genome will be produced at Yale and elsewhere, and will need to be stored in an organized fashion so it can be retrieved and displayed easily. Many different types of data are being produced which characterize the genome at different levels of detail. This information is typically uncertain and incomplete, and data at one level may interact with (reinforce, complement or contradict) data at another level. The routines that retrieve this data should adapt intelligently to the uncertainty, incompleteness and interdependencies of the data. Representation of Physical Mapping Data. The laboratories of Drs. Kenneth Kidd, David Ward, Sherman Weissman, Frank Ruddle, Stephen Reeders, and Allen Bale are currently producing large quantities of physical mapping data, including large scale restriction maps, sets of overlapping cosmids, mapping of yeast artificial chromosomes, and in situ hybridization data. We will develop and refine an appropriate representation for this data. We will focus particularly on how best to represent the inherent uncertainties and interdependencies in the data, and consensus conclusions based on the data. Intelligent Retrieval and Concise Display of the Data. We will develop routines that incorporate several types of knowledge to allow intelligent retrieval of the data. We will also develop routines to display the data graphically at different levels of detail, including the construction of gene maps. Although the data in the database itself will be complex and interrelated, the display routines will be designed to present the data clearly, at an appropriate level of detail. For example, a map of a chromosome or of a large or small chromosomal region will be concise and easily understood.