IXDB, an integrated database for genomic data of the Human X Chromosome

The generation of physical maps at the chromosomal scale and at a resolution suucient for large scale sequencing requires more data, experiments and resources than can be produced by a single research group. This is reeected by the fact that most chromosomes are informally split into regions of research interest. These are then often exclusively tackled by a small number of groups, following certain specialized research interests such as the search for a speciic gene. In addition, many interesting questions other than those aimed at genome sequencing require an overall overview, such as the determination of chromosome structure conservation across species. Such questions are currently very tiresome and error-prone to approach, as no single-point and comprehensive interface to all the diierent datasets is available ((1]). We have developed IXDB, a database focused on the human X chromosome, to assist our mapping projects. IXDB has a relational datamodel and covers the eld of mapping information, with growing extensions towards genomic sequences and a more detailed gene representation. IXDB is designed as an integrated database, which means that it includes and tightly integrate data from other public database. It aims to ultimately serve as the above described single-point interface to mapping data of the human X chromosome. It currently contains information from many diierent sources, including GDB, OMIM, the literature or the EC X Transcript Map Consortium. Around the core database lie three applications implemented in TCL/TK. These include: 1.) a public WWW interface ((2]) that allows for object search by name, keyword or map location and includes a graphical map representation via a JAVA applet; 2.) a map display tool that produces high-quality maps suitable for presentation; and 3.) MAPDIF, a tool for a mathematically sound comparison of marker maps which allows the measurement of the \distance" between two maps. According to 3], IXDB follows a data warehouse approach to achieve data integration. All data is stored locally. Advantages of this strategy are high availability, tight data integration and the potential to formulate arbitrary queries, making data mining strategies possible. The main disadvantages lie in the diiculty to maintain the data up-to-date, and the relatively high disk