Operon prediction in Agrobacterium tumefaciens C58

Project highlights:

The following data were generated using prediction software developed by Westover et al.

• Westover, B.P., Buhler, J.D., Sonnenburg, J.L., & Gordon, J.I. (2005) Operon prediction without a training set. Bioinform. 21, 880-888.

The scripts described above were generously corrected and updated by Nalvo Almeida Jr. of the Federal University of Mato Grosso do Sul, Brazil. Nalvo is currently on sabbatical in the Setubal lab at the Virginia Bioinformatics Institute.

The predictions below were made using the following parameters:

• Genbank files NC_003062, NC_003063, NC_003064, NC_003065 retrieved on 7 November, 2007
• Threshold beta = 0.7
• Training set: B. hensulae (NC_005956), B. japonicum 110 (NC_004463), B. pertussis (NC_002929), B. suis 1330_c1 (NC_004310), B. suis 1330_c2 (NC_004311), C. crescentus (NC_002696), M. loti maff 303 (NC_002678), N. hamburgensis x1 (NC_007964), R. etli cfn42 (NC_007761), R. leguminosarum vic3841 (NC_008380), S. meliloti 1021 (NC_003047). Only a subset of these genomes were selected by the program for the threshold given.

These predictions are based on comparisons with closely related genomes. Additional predictions using a larger and more distantly related training set are in progress. Please see the paper by Westover et. al for details relating to the following data.

• Operon predictions (6 January 08)