coords always written in order seq_id, seq_start, seq_end [, seq_strand]
added register_region() to map_coordinates_to_rawcontig()
some pod tidying
changed some internal method names (e.g. 'freeze' to 'sort')
sql cleaned
bug squashed in Mapper.pm in gap regions
added more sanity checking of arguments in map_coordinates_to_blah()

Possibility to add of RefSeq entries to get gene description with precedence on SPTREMBL,
and SWISSPROT having precedence on RefSeq.
Then ranking descriptions using the similarity measure used in establishing the mapping.
Manage now to load gene descriptions file to the corresponding Ensembl core database.

Integration of select query (from sp-mapping-dump.sh) in gene-description.pl.
Next step, integrate sp-description.pl in gene-description.pl, so gene description can be manage through one command :)

Then identified bug in Accession Number parsing, when in more than one line for an entry, script was getting the first
AC of the last parsed line. But we want the first AC of the first parsed line. Fixed now. Added possibility of parsing

This  script is now deprecated. It has been integrated in gene-descriptions.pl

Now does two different mapping, one for the prediction and one for the known genes... it will allow us to be more free to tickle the parameters

added peptide ---> contig coordinate mapping on Transcript. Not tested. Mainly to make sure I shift the code off my laptop

same fixes as in branch-4: get_Genes_by_Type treats nicely the supp_evidence and invert() preserves the db object in contig

Not much gain to see so far, all alignments look ungapped.
Probably still buggy.

not ready yet.

Adaptor to store in db and retrieve.