Changed the big picture to show coords on a per-nucleotide basis rather than per-peptide, which fixes a bug where there were sometimes gaps between adjacent exons and introns.

Fixed a bug where CDS/UTR boundaries were calculated incorrectly for exons on the reverse strand. Also fixed a bug where exon view was not redrawing on window resize

Made the big picture scrollable, which fixes some issues where the big picture section can get too large when exons are bumped or the grid scale is changed

Added flag for highlighting polyA signals in the reference sequence. Made higlighting of special bases in the reference sequence more generic

Added poly_site and polya_signal_sequence GFF types. Consolidated some code that calculates the full display range of an MSP including any unaligned portions of the match sequence

Fixed a bug where pfetch was being called multiple times when starting dotter. Changed the client name sent to pfetch from 'acedb' to the application name

Added bounds-checking on manual dotter coords. Changed several dialogs so that they are not closed if there is an error with the data entered.

Dotter: replaced messalloc/messfree by g_malloc/g_free. Removed some unused items from dotter_.h and removed use of dotter_.h from belvu because it doesn't seem to contain anything that is being used.

Added a prefix to the 'organism' data field so that a narrow column width can be used that displays just the prefix

Fixed a couple of bugs with squash matches: go-to-next was highlighting the wrong frame for protein matches; MSPs for both strands were being added to each tree.

Removed the need for the -F command line option when a single input file is used. If blixem is called with one input file, it is assumed to contain both alignment data and the reference sequence. If it is called with two files, the first is assumed to contain the reference sequence and the second the alignment data. The -F option has been left in for now for backwards compatibility.

Fixed a couple of bugs with the GFF parser: sequence name was not being read from the header; fixed a problem with constructing introns

Added two status bars: one for messages and one for feeding back info about the currently moused-over item. Fixed a bug with parsing peptide sequence data from seqbl files. Fixed some bugs with negative offset: some termini were being shown as splice sites; go-to-next match was sometimes messing up.

Added support for grouping exons by parent transcript ID (i.e. as passed in GFF format). Added full support for CDS/UTR types. Calculate the reading frame, using the given phase, if applicable. Changed score and ID to be decimal values rather than integers and allow grid scale to show fractions of an integer per cell. Reduced exon size and padding when the exon view is bumped. Added GT-AG to list of canonical splice sites. Removed incorrect splice-site highlighting which was being shown at the termini.

Removed a spurious popup message that happens when pfetching full entries on startup fails (because pfetching just the fasta sequence may still work fine)