Note! This is currently not working and needs some attention to work with any of the servers for which plugins were written.
There is a frequent need to apply the same set of prediction and annotation tools to a large batch of sequences. Such tools may reside locally or on remote servers accessed over the web. In order to create a meta-tool able to dispatch one or more sequences to assorted annotation/prediction services it is necessary to define a consistent format for the data required by such services and for the annotations which they provide.
We have determined that the output can be described using one of 6 forms of data: numeric or textual annotation of residues, domains (residue ranges) or whole sequences. A tool may produce a combination of such outputs. With this in mind, an XML data-type definition (DTD) was designed to store the output of any server (`Automated Protein Annotation Tool Markup Language', APATML).
APAT lets you write simple wrappers for annotation servers which then generate APATML as output. Our display program will then format that as HTML (including colouring of residues and graphs) for you to view, or you may write your own programs to extract and analyze the data.
A paper is available describing the system has been published in Bioinformatics, 22,291-296.
APAT was written by PhD student S.V.V. Deevi funded by The Felix Trust at Reading University.
APAT is freely available for use by not-for-profit organisations. Commercial use is not permitted without express permission from the author. It may not be distributed without the author's permission, but must be obtained from this site.