There are two complications: first, different organisms are studied with different experiments. In C. elegans, for example, there is a rich history of classical genetics, but not many researchers study quantitative loci or use RFLPs. However, people studying conifers make heavy use of RFLPs while anyone involved in crop improvement is interested in quantitative traits. The result is that distinct kinds of data need to be represented in each case. Second, research communities may have idiosyncratic traditions that affect the way data is structured. A community might agree to assign each laboratory a "laboratory code" and use this scheme to identify responsibilities (parts of the genome being sequenced or mapped; allowable symbols for novel mutations, etc.). In this case a Laboratory class that connects to sequences or genes might be important.
Given that reconfiguration is typically required for a new database, you will find it useful to start with a working set of models -- for example, the C. elegans models -- and modify them as needed, . This is the approach every new group has used thus far. Installing and examining one or more "alien" databases is an excellent way to shop for useful ideas. Nearly all database curators are willing to provide a copy of their most recent models on request and will answer questions about them. See the ACEDB FAQ for curator names and contact information.
It is also possible you are not interested in biology at all and intend to use ACEDB for an unrelated purpose. The database can be configured in a "generic" state (Appendix 1) and models can be designed from the ground up. The result will be quite suitable for handling diverse text-based information.
Back to Table of Contents