EMBOSS: Project Meeting (Fri 13th March 09)

Attendees

EBI: Peter Rice, Alan Bleasby, Jon Ison, Mahmut Uludag
Sanger:
Visitors:
Apologies:

1. Minutes of the last meeting

Minutes of the meeting of 2nd March 2009 are here.

This meeting was moved because Peter is away on Monday 16th.

2. Software Development

2.1 Applications

Peter has modified the feature types used by applications to define them using terms from the Sequence Ontology. Previous releases often used the artificial name "hit" which was represented as an undefined sequence feature. The SO terms produce better results when they appear in the output, especially in DASGFF for DAS services, and in GFF3 which is the default feature output. Report outputs are unaffected.

2.2 Libraries

Peter has updated the feature table internals for nucleotide sequences to implement the changes in the EMBL/GenBank feature table v8.0. Non-coding RNA features have been rationalised to use a single ncRNA feature key with a qualifier to identify the ncRNA type. Sequence Ontology equivalents have been identified for immunoglobulin features. Only the S_Region had no exact equivalent and is represented by an artificial internal identifier derived from SO:0000301.

Peter has further updated the DASGFF output to use a unique feature identifier. This was required to correctly display results in the Ensembl browser.

Peter will investigate the possibilities for displaying sequence-based graphical output in DAS clients.

2.3 SoapLab

Mahmut has corrected some issues with concurrent executions and threads. Service error reports were generated by the test process taking longer than the timeout period.

2.4 Other

Peter and Mahmut attended a DAS workshop earlier this week. New features planned for DAS 1.6 (the current version is 1.53E) will be useful for our EMBOSS-based DAS services

Mahmut will try using the proserver Perl-based DAS server to replace myDAS java for the EMBOSS DAS services.

3. Administration

Alan has set up the EMBOSS Open-Bio wiki. A small EMBOSS logo has been generated from the first letter of the standard logo. Word filtering is in place. Bibliographic references can be linked by PubMed IDs, etc. Text and images are licensed under GNU FDL and creative commons.

Anonymous editing is allowed, but can be restricted to users with accounts. External URLs generate an automatic check. Image uploads are allowed, preferably as GIF or JPG. The top page is protected from anonymous editing. Backups are to be done by the Open Bio team. Google analytics are turned off.

4. Documentation and Training

4.1 Books

Jon reported that the CUP technical staff cannot process DocBook XML directly, and prefer a MS-Word version. Jon will investigate options for conversion of trial chapters via other stylesheets (e.g. LateX, PDF) as conversion via HTML can lead to bloated MS-Word documents. Some mechanism will be needed to import corrections from CUP.

4.2 Training

The Madrid course clashes with other commitments and will have to move later, possibly June.

5. User queries and answers

No new items

6. AOB

None.

7. Date Of Next Meeting

The next meeting is also in the downstairs meeting room by the new training facility.