GenProp2023: Merged Steps?

[LeeBergstrand]

GenProp2023 has the following listed steps:

--
SN  1
ID  Caspase-2
DN  Caspase-2 (EC:3.4.22.55)
RQ  1
EV  IPR035702; PTHR10454:SF151; sufficient;
TG  GO:0006919;
ID  Granzyme B
DN  Granzyme B (EC:3.4.21.79)
RQ  1
EV  -; PTHR24271:SF38; sufficient;
TG  GO:1900740;
TG  GO:0004252;
--
SN  2
ID  BH3-interacting domain death agonist
DN  BH3-interacting domain death agonist
RQ  1
EV  IPR010479; PF06393; sufficient;
TG  GO:0008625;
--
SN  3
ID  Apoptosome
RQ  1
EV  GenProp2021;
--
SN  4
ID  Extrinsic initiator caspases
RQ  1
EV  GenProp2018;

Are the are Caspase-2 and Granzyme B in the same step? Should they be separate steps?

[LeeBergstrand]

Also the EV -; PTHR24271:SF38; sufficient; the first link out has -;

[LornaMGnify]

Thanks for your comments! If I remember correctly (I didn't actually build this property) the first pathway step can be carried out either by Caspase-2 or Granzyme B. These are not two separate steps, but it is a slightly different situation to there being multiple different lines of evidence for a single enzyme. We are still deciding how to best represent this in the DESC file. GenProp2023 is still being edited (it is not yet ready to be made public as you can see from the public=0 in the status file) partly for this reason. The other reason that it is not public is that some of the evidence is not yet integrated into InterPro - hence the -; you mention, where the InterPro accession will be. Hope that helps. Out of interest, what is your interest in this data? Are you using it, or planning to?

[LeeBergstrand]

@happy-lorna Thanks for the quick reply!

For my MSc. thesis project I am building a piece of bioinformatics software which builds on top of Genome Properties. Through some research I noticed that Interpro has taken over their development from JCVI. The software I'm building is non-overlapping with Interpro's genome-properties-viewer, though it is also a web application.

If your group is interested in learning more I could perhaps send an email (GenProp@ebi.ac.uk ?) with more details about the project.

At the moment, I am prototyping a Python library for parsing genome property flat files into Python objects. The code for it can be can be seen here (open sourced under APLv2). I'm working against the flat file found in your latest test release (here), however, I decided to test it with the non-public files as well.

Cheers!

[LeeBergstrand]

So the ID tag would be the differentiator between the two evidences?

[LornaMGnify]

@LeeBergstrand Please do send us an email at GenProp@ebi.ac.uk to let us know some more about your project. I can then let you know a bit more about what we are woking on too!

As for the merged steps question - we have not yet decided how best to represent this situation. It may make more sense to provide a more generic ID for the step and then list the two enzymes as alternate evidences for the step.

[LeeBergstrand]

@happy-lorna Will do.

[LeeBergstrand]

@happy-lorna Email Sent.