WWW-ORIENTED BIOINFORMATICS SUPPORT FOR CONSORTIUM-BASED GENOME PROJECTS

Bioinformatics Laboratory, State University of Campinas, Campinas, SP, Brazil, Box 6176, 13083-970, URL: www.lbi.dcc.unicamp.br, vox: +55 19 788 5849, fax: +55 19 788 5847

The Laboratory of Bioinformatics (LBI) of the University of Campinas, Brazil, provides services to 3 genome projects in Brazil: (1) whole genome of the bacterium Xylella fastidiosa, a citrus trees pathogen; (2) whole genome of the bacterium Xanthomonas citri, another citrus trees pathogen; and (3) EST of sugar cane. All of these projects are consortium-based, meaning that sequencing is done by several labs from different institutions, geographically scattered over a wide area. The number of labs working on each project ranges from 12 to 34. The consortium approach was chosen because it allows pooling of resources and provides training (in each project there is a mix-up of experienced and inexperienced labs). However it depends crucially on a bioinformatics central facility. The LBI is such a facility, and thanks to its role has developed considerable expertise in supporting lab consortia. The main services provided are related to the management of a project-specific sequence database, where "database" is taken in its broadest possible sense. A sample of these services is: chromatogram submission over the WWW and processing; DNA assembly for whole genome projects; preliminary ORF annotation and other "gene mining" resources; and EST clustering. Our methodology has centered on the use of publicly available software whenever possible. Even so, we have carried out considerable in-house perl-based software development, mainly related with the sequence submission pipeline. All our services use the WWW as their interface and they have been used intensively and successfully by all the consortium laboratories. The LBI is staffed by around 5 people. It is relatively small considering the size of the projects it supports. LBI is supported by FAPESP, Fundecitrus, and Copersucar.