About ATIDB's Brassica features


ATIDB now displays Brassica sequences positioned by similarity to the Arabidopsis TIGR v5 genome sequence, thus integrated with the TAIR v6 annotation and a wide range of functional genomic data (courtesy of Sean Walsh of the John Innes Centre). Clicking on any Brassica feature pops up a window from which a realtime multiple alignment can be launched for all Brassica sequences over that region. Please note that you must have Javascript and Java enabled and allow popups from brassica.bbsrc.ac.uk for this to work.

Updated 13th October 2006: a total of 808,039 GSS sequences (comprising B. rapa BAC end reads and B. oleracea shotgun sequences from TIGR/CSHL) and 127,537 ESTs were extracted from EMBL/GenBank and compared against the TIGR v5/TAIR v6 Arabidopsis pseudochromosomes using WU-BLASTN, with an E-value cut-off of 1E-10. The best scoring HSP(s) were selected for each query sequence, the absolute chromosome coordinates recorded as a feature in GFF format and loaded into the MySQL database driving the GBrowse display. Where there were tied scores for the most significant HSPs each candidate mapping is now also recorded in the database. The significance of each BLAST hit is colour-coded as shown. For BAC clones (see below), the coding refers to minimum identities at both ends.

HSP identity <60%  60-70%70-80%>80%   
Brassica feature

53% of the GSSs and 90% of the ESTs could be "mapped" in this way, but be aware that these mappings potentially conflate paralogous copies (revealed by the colour-coding scheme and by Jalview's tree display). The object is to offer a route to possible Brassica homologues for a given gene, not to display chromosome synteny as such.However, the latter may be partly addressed by analysis of a subset of the GSSs, originating from 200,031 BAC ends of the Brassica rapa KBr libraries. We post-processed those mappings to identify pairs of BAC ends giving hits to Arabidopsis consistent with microsynteny up to 500 kb. About 14% (11,709) of the BAC clones analysed were mappable and satisfied these criteria. These are now displayed, using the same colour coding, along with "orphaned" BAC end hits. The data, subject to continual revision, are summarised here.






For help, comments and bug reports, please contact Martin Trick