OVERVIEW     ONTOLOGIES     WEBSERVICES     BIOMOBY     MOBY CLIENTS  

The BioMOBY project

  Overview
 
  This course aims to give a general overview about BioMOBY.
BioMoby is a project to develop a Web services architecture for bioinformatics. It aims to explore various methodologies for biological data representation, distribution, and discovery.
We will introduce the architecture of the MOBY framework.
Then we will present various clients that can facilitate the use of MOBY Web services and that can enact data workflows.

BioMOBY course presentation

  Exercises
 
  These exercises aim to illustrate the use of the Biomoby infrastructure to execute Bioinformatics tasks.

Exercise 1:
You want to analyse a protein, which UniProt identifier is the following, "src_human". Using Mowserv and Remora, retrieve the protein sequence from UniProt, then search using a Blastp service against the Swiss-Prot database, finally, filter the best hits, those with e-value lower than 1e-10.

Using Remora

1. Go to Mowserv website, http://lipm-bioinfo.toulouse.inra.fr/remora//cgi/index.cgi?FORM=1.

2. Select the "Inab" registry, unselect the "mobycentral" registry.

3. Validate the registry you have chosen by clicking "Start session".

4. Select the input "Data Type", from the "full list" to be "Object".

5. Select "UniProt" as the Namespace of the input protein identifier.

5. Validate the input DataType by clicking "Dive into the sea" button.

6. In the list of proposed Web services, select "getAminoAcidSequence".

7. Validate your selection by clicking "Add Service" in the bottom of the page.

8. Now you can see a graph of the workflow that is being implemented.

9. Click on the output object ("sequence").

10. In the list of proposed Web services, select "runWUBlastp" from "inb.bsc.es" server, as well as "fromGenericSequenceToFASTA", and validate your services selection.

11. Your workflow is done. Go to Step 2, i.e. go to the workflow configuration.

12. Click on the "Object" (displayed in Red) to fill the input.

13. Enter in the "Id" field, "src_human", then click "Save".

14. Calibrate the Blast search. To do so, click on the "runWUlastp" service.

15. Select the database to be "Swiss-Prot".

16. Set the filtering to "seg+xnu".

17. Select the word_size to be 6.

18. Validate your new set of parameter values by clicking "Save parameters".

19. The worklow is now ready to be executed, go to step 3, "Workflow Submission".

20. Enter your email address to be notified when the worklfow execution is done, and a workflow identification name.

21. Check in your email when you are notified that the workflow execution is complete.

22. Check the results.

Using Mowserv

1. Go to Mowserv website, http://www.inab.org/MOWServ/.

2. Click "Expand all" in the left menu

3. You get a list of services, that can be browsed through three catalogues, "DataTypes", types of "Services" and "Namespaces".

3. Select "getAminoAcidSequence" service, for example from the "UniProt" namespace categorie.

4. First is to create the input object, by clicking "Create" button.

5. Fill the "Id" box, with our protein identifier, i.e. "src_human".

6. Fill the "Namespace" box with "UniProt".

7. Then Click "submit" to validate your input object.

8. Now you can execute the "getAminoAcidSequence" service, by clicking the "submit" button.

9. The service is now being executed, go to "User Objects" tab to see the status of the execution.

10. Click on the "AminoAcidSequence" resulting object to connect it to the next service.

11. You get a list of services that can take an AminoAcidSequence, select "runNCBIBlastp" from "inb.bsc.es" server.

12. Click "Execute", you get a page with parameters settings.

13. Select the database to be "Swiss-Prot".

14. Set the matrix to be "BLOSUM80".

15. Set the filtering to "true".

16. Now execute the service by clicking "Submit".

17. Redo the same process with the next service, called "getBestHitsFromBlast".

18. Select a Threshold of "0.00001".