EMBOSS: Project Meeting (Mon 23rd May 11)


EBI: Peter Rice, Mahmut Uludag, Michael Schuster
Apologies: Alan Bleasby, Jon Ison,

1. Minutes of the last meeting

Minutes of the meeting of 16th May 2011 are here.

2. Maintenance etc.

2.1 Applications

Peter fixed tcode to report results for non-coding nonsense data by expanding the y-axis to cover the threshold values.

2.2 Libraries

Peter fixed graphics applications to correctly report separate XY plots for multiple input sequences. The application example results need to be carefully checked before the release.


Peter has fixed tests for dbigcg and ontoget in the Visual Studio builds.

2.4 SoapLab

Martin has prepared a new test distribution which Mahmut has tested on the London Data Centres. Some issues with message part attribute names need to be discussed. The latest SoapLab should be live soon.

3. New developments

3.1 Access methods

Mahmut has updated DAS and wsdbfetch access. As dbfetch returns metadata in the same form as wsdbfetch we can auto-generate the server cache files. Dbfetch access provides the same data resources as wsdbfetch without the Axis2C dependency.

Mahmut has extended CHADO access with some example GFF files. Sequence-region lines are added to the output.

FlyBase CHADO access is under development, and not yet included in the QA tests. It should be usable by the release.

Michael noted the need to map Ensembl and GFF features. In Ensembl all features are chromosome based but can be mapped to other coordinate systems.

Michael has updated to Ensembl 61 and committed the code changes. Server cache files are committed for Ensembl and Ensembl genomes with 100-400 databases in showserver plus alias names. The databases are dumped in alphabetical order, as are the dbalias names.

Ensembl 62 is now out. Most changes are in variations. Some more testing is needed for the API.

Michael would like help in assigning EDAM links to Ensembl databases.

Mahmut looked into MRS support. The support for formats is not good but there is help on the query interface.

3.2 New applications

Peter has extended textget to report text from sequence or obo entries where the appropriate format name has a text parser that can find the end of an entry. Some cleanup is needed to remove attempts to read these entries as sequence or obo directly.

Applications that create server cache files need a standard naming convention for the release. They should also have a standard interface to ask for server name or other details.

Mahmut suggested adding comment lines to server cache files to explain how the file was created.

Peter proposed adding a variation object to ACD and AJAX to make use of the Ensembl API and for reading/writing VCF file format.

Mahmut suggested explicit support for paired-end reads for NGS data.

4. Administration

Peter would like to update the EMBOSS version number to "" on all systems so that QA tests give consistent results. This failed on Unix as the extra digit is not valid for the build utilities. Peter will consult Alan on the best approach.

Mahmut is interested in installing Ubuntu as a windows application using Wubi. Michael offered to demonstrate the use of VirtualBox by Ensembl to test Unix distributions under Windows.

5. Documentation and Training


6. User queries and answers

All done.

7. AOB


8. Date Of Next Meeting

Next week is a public holiday. The next EMBOSS meeting will be on Monday 6th June.