EMBOSS: Project Meeting (Nov 1st 1999)


Attendees

Sanger Centre: Peter Rice, Ian Longden
HGMP: Alan Bleasby, Gary Williams, Mark Faller, Val Curwen
EBI:
Visitor: Keith Vass (Glasgow)

1. Matters Arising

None

2. General progress on release 0.0.4

The major need is for additional database indexing formats. Peter is writing code to index Blast1, Blast2, GCG and NBRF formats with EMBLCD/Staden style index files.

EMBOSS can already read the index files directly from the EMBL database CD-ROM distribution for EMBL and SwissProt.

Documentation on database definitions is needed. Peter will write a comprehensive guide once the new formats are implemented.

Sequence type testing has occasional problems, for example aligning two sequences needs a decision on the type of the first sequence. A length limit will be set of testing DNA or protein sequence type so that any sequence longer than, for example, 50 characters and containing only DNA base codes will be set to nucleotide and not protein. In testing a protein sequence this would result in a warning message only for now. The previous problem of short SwissProt entries should be resolved by using the database type and sequence format where this is informative.

Alan has a new application "plotorf" which plots open reading frame information.

Alan modified the output of MSF format to add missing dots on the checksum line. Peter will check on the general handling of MSF format.

Alan has generated an RPM file for EMBOSS available from the HGMP FTP server.

Implementing a new release number, which should be done soon because of the number of changes, is a simple change to the automake file.

"emma" had some problems caused by waiting for a child process when there were other child processes open (e.g. reading a database with SRS). The code in "emma" now explicitly waits for the "clustalw" process to exit.

"tmap" had a bug when reading single sequences which has been fixed. This was a bug in the original version.

Alan has updated "syco" and "chips".

Ian is working on Will Gilbert's MSE editor. Most of the operations are working. This application will need to prompt the user for additional sequences, but will use the ACD file to startup.

3. CCP11 Course

Val has prepared the exercises. The practical session will be run by Peter with help from Alan, using HGMP accounts for the students.

4. Interfaces

Thon is investigating Pise for a Web interface and disGUIse for X11. Mark will work on a prototype of EMBOSS applications under disGUIse. Alan will look into Pise. Peter is looking into SRS6 where similar issues arise.

5. Any Other Business

Alan has implemented a Birkbeck Template Library (BTL) plugin for EMBOSS in C++ with 2 example programs. The ACD files are included in the EMBOSS distribution for simplicity. The BTL library and applications are in a separate top level directory.

6. Next meeting

Next meeting Monday 8th November, 11:00am, usual place.


Peter Rice, Informatics Division, The Sanger Centre, Wellcome Trust Genome Campus, Hinxton Hall, Cambridge, CB10 1SA, UK.