Overview of approach for analyzing prefinished genomic sequence data. Prefinished sequences for individual clones are posted on a public access ftp site. By use of either a web browser or ftp client, this data can be downloaded and analyzed by two main classes of programs: (1) those that compare the sequence with existing databases (e.g., BLAST, PowerBLAST; left) and (2) those that predict or identify features intrinsic to the sequence (e.g., GRAIL, MZEF, GenScan, SEG; right). The first class of programs can be used with public (e.g., dbEST, dbSTS, nr) as well as locally created databases (DB), such as specialized collections of cDNA sequences. The results generated with both classes of programs are collected as annotations within a common Sequin record. Several filter programs (indicated in yellow) must be used to take the textual output from some analysis programs and convert it into ASN.1 format. Such filters are relatively easy to implement and make this system usable with most (if not all) sequence analysis programs. This conversion allows all of the analyses to be stored in a computable format that can be readily updated as more complete versions of the sequence become available (see Fig. 3). The spatial relationships among all annotations can be edited and viewed by either Sequin or Musk. See Table 1 for additional details about the various programs and resources.
