v1 - 12/17/2005 
  This dataset contained 5707 repeats obtained from the MIPS repeats,
  137 repeats from the unirepeats set that didn't match the MIPS set,
  and 7 sequences from Genbank from known tomato repeats (TGR
  repeats). The unirepeat set were derived from bac ends using
  RepeatScout

v2 - 08/06/2008
  A combination of the PRI set from 2008-07-23 and a bac-end derived
  repeat set (unirepeats)

v3 - 08/12/2008
  A combination of the PRI set from 2008-08-12 and a bac_end derived
  repeat set (unirepeats)

v4 - 11/04/2008
  Combination of RepBase viridiplantae and PRI set. The unirepeat set
  was removed because it was too sensitive and interfered with
  annotation (possibly contained large gene family sequences as well).

draft v5 - 02/25/2010
  Combination of repeats of v4 plus the results of the analysis of 
  tomato genome scaffolds v.1.03 with RepeatScout. These results were
  filtered and all the sequences with some gene homology (matches with
  Arabidopsis protein dataset using blastx and e-value < 1e-10), low
  complexity (analyzed by nseg program) or tandem simple repeats (
  analyzed by trf program) were removed from the dataset.

v5 - 03/23/2010
  Combination of repeats of v4 plus the results of the repeat analysis of 
  tomato genome scaffolds v.1.03 with RepeatScout. These results were
  filtered and all the sequences with some gene homology (matches with
  Arabidopsis protein dataset using blastx and e-value < 1e-10), low
  complexity (analyzed by nseg program) or tandem simple repeats (
  analyzed by trf program) were removed from the dataset. Also the 
  redundant sequences between repeat v4 and RepeatScout analysis were
  removed. Finally a RepeatMasker analysis were made using this dataset.
  All the sequences repeated less than 10 times were removed.