This is the canonical repository for BAC sequences produced as part of
the International Tomato Genome Sequencing Project.  For more
information about the project, see
http://www.sgn.cornell.edu/about/tomato_sequencing.pl .  For the
latest assembled tomato genome contigs, see
ftp://ftp.sgn.cornell.edu/tomato_genome/contigs/ .


CONTENTS OF THIS DIRECTORY

* bacs.v*.seq
   FASTA-formatted sequences of all submitted BACs
* bacs_repeatmasked.v*.seq
   same as above, but with repetitive sequences masked with
   RepeatMasker, see
   http://sgn.cornell.edu/gbrowse/gbrowse/tomato_bacs/?help=citations#RepeatMasker
* bacs_accessions.v*.txt
   list of GenBank accessions for BAC sequences
* finished_bacs.v*.seq
   sequences of only finished BACs
* finished_bacs_repeatmasked.v*.seq
   repeat-masked sequences of only finished BACs
* finished_bacs_accessions.v*.txt
   list of GenBank accessions for finished BAC sequences
* validate_submission.pl
   perl script used by sequencing centers to check the format of a BAC
   tar.gz file before uploading.
* 2764_finished_bacs_htgs3.fas
   Includes 2764 non-redundant phase 3 BACs from finished_bacs_accessions.v676.txt 
   and NCBI. Used for SL3.0.

chr##/        - materials for BACs on chromosome ##
nonstandard/  - contains additional data that do not conform to formatting
                guidelines


A NOTE ON SEQUENCE AND FILE VERSIONING

There are two systems of version numbering in place in the SGN BAC
repository, one for files, and one for BAC sequences.

File versions track the history of each file in the BAC repository.
When a change is made to a file, its version number is incremented and
a copy of the previous version of the file is stored in a subdirectory
called old/.  For example, if a new version of the file
C01HBa0216G16.1.v2.tar.gz is published, the new file will be called
C01HBa0216G16.1.v3.tar.gz and the old version of the file will be
archived in old/C01HBa0216G16.1.v2.tar.gz.<timestamp>.  Using this
system, a full version history is kept for all files in the BAC
repository.

Sequence versions track the history of each BAC sequence.  It is
incremented only when a change is made to the sequence.  For example,
in C01HBa0216G16.1.v2.seq, the first line might be '>C01HBa0216G16.1',
indicating that this is the first submitted version of the sequence
for the BAC C01HBa0216G16.  If an update to this BAC is submitted with
a different sequence, the seqquence version for the BAC would be
incremented, resulting in a file named C01HBa0216G16.2.v1.seq, with a
first line of '>C01HBa0216G16.2'.

On the other hand, the sequence version will NOT change if an update
is made that does not change the sequence, such as changing
annotations or adding supporting data.

For BACs whose sequence is still in pieces, a given sequence version
applies to the set of pieces, not each individual piece.  For example,
suppose that C01HBa0088L02 is first submitted to SGN in three pieces.
Its sequence file, C01HBa0088L02.1.v1.seq, will contain the fragments,
'>C01HBa0088L02.1-1', '>C01HBa0088L02.1-2', and '>C01HBa0088L02.1-3'.
Then, if an update is made changing the sequence of one fragment, the
sequence version is changed for all the fragments, making:
'>C01HBa0088L02.2-1', '>C01HBa0088L02.2-2', and '>C01HBa0088L02.2-3'.
If a third update then has a single, finished sequence, the identifier
for that would then be '>C01HBa0088L02.3', since it's the third
version of this BAC's sequence.


--
 Please note that this is a working document, subject to change at
 any time without notice.