EMBOSS: Project Meeting (Fri 14th January 2005)


RFCGR/HGMP: Alan Bleasby, Jon Ison, Gary Williams,
Sanger: Tim Carver
EBI: Peter Rice, Lisa Mullan


1. Minutes of the last meeting

Minutes of the meeting of 12th November 2004 are here

2. Software Development

2.1 Database indexing

Alan described the embindex source code which has grown substantially.

Keywords will be implemented as a tree of trees. Aim to add in time for the next release for index and retrieval. Updating and deleting of keywords will follow.

2.2 Alignment formats

Peter reported on a major review and bugfix of sequence alignment output formats, started by a bug report on reversed sequence numbering.

2.3 diffseq update

Gary has added an option for diffseq to work with a global rather than a local alignment.

2.4 Antigenic Regions

Lisa reported a request from Phil Cunningham at King's College London, who had compared antigenic with the GCG program peptidestructure on selected SwissProt entries and found the GCG output may be more accurate (though not perfect) but is very hard to read.

An EMBOSS version of the Jamieson-Wolf algorithm would be very useful. Peter noted that he has had a similar request from Schering.

2.5 Domainatrix

Jon has updated all applications, and incorporated Waqas's and Mike's code. ACD files and file formats are all updated. Some revisions are still needed for the PDB parsing paper, especially for better ligand detection where PDB file markup is poor.

Next month the domainatrix package may be split into 3 smaller parts.

3. Administration

3.1 Release 2.10.0

Alan will add the indexing code.

Peter will complete testing of alignment formats with reversed DNA sequences and ranges.

Jemboss should use the embassy directory for documentation of EMBASSY programs.

4. Documentation & Training

4.1 Books

Docbook tags are proving too complex. General preference was to use a Word template instead. The deadline for the books is unclear, given the RFCGR closure.

4.2 Code documentation

Peter is adding category markup for EFUNC, with the aim of validating categories and ensuring they appear in the correct "section" of the source file. This will be a gradual process, starting with string functions.

4.3 Sourceforge website

The cygwin page is incorrect and needs updating. Alan has the details.

5. User queries and answers

Gary's current list reviewed. Most new issues were already answered.

6. AOB

7. Date Of Next Meeting

Next meeting at 9.30 on January 28, 2005,