From NCBO Wiki
Jump to: navigation, search

Rough minutes from Phenote workgin group meeting with Nescent 5th June 2007

Jim Balhoff:

1. How to apply EQ to evolutionary characters

Coming to method closer to that being used for Mutant phenotypes than thought would at first.

-> phenote interface style OK

2. workflows

Starting with a a paper -> set of specimens. They will often want to make statements about subsets

Might want to make say one statement about catfish and one about salmon...


ONe thing that would be extreemly useful for us is to do some upfront setup work.. There are lots of things that will pertain to everything you will record for one paper.

You want to be able to record in advance what the species are

Paula - the examplar approach - we are envisioing setup starting with listing examplars.

So - you put in your speciment list first.

Since you'll be using a taxonomy - if you want to have a shortcut where you can apply a statement to all members of a group.

But some issue of how this is linked - best linked to exemplar species. It is a hypothesis that something applies to all because it applies to exemplars.

MA: Each order will I presume have a set of EQ pairs which differentiate - and all the way down through orders, families, species etc

Then all you would have to enter is the differentii for species specificity.

Paula says - this is putting conclusions first - need to keep data linked to observation.

Jim Balhoff

Bringing it back to practicalities:

What we need is a shortcut that allows you to choose a taxonomic grouping and then automatically generate the complete set of statements for examplars used in that paper for that grouping.


Two more things

1. PATO:

JIM - attribute and value slims - has anyone else found a use for these? Presumably these will need to be maintained. We still think they will be valuable to us.

Paula - I would think that the phenotypic data should require these slims. DS: Think most likely to be useful for front end - analysis tools for PO annotated data.

2. Character Matrices:

Character matrix view of the data needs to be easily accessible. Press a button and convert phenote table view to character matrix (presumably using attributes slim).

Jim has written something for this - only works in one direcetion (phenote -> character matrix ) right now.

[MA - says something about spider phenotype ontology - Nicole working with these people...]

BIRN - interested in charcter matrices. - according to Nicole just interested in this as a particular view.

"If we are to adopt phenote as a basis for our editor, we would certainly be in a position to play a major driving role".

Mark asked - Is this interest in phenote web based:

A: Not sure.

What are your goals over the next few months:


1. config GUI

2. Tab system - one tab informs another - this will work for genotypes.

A: That sounds really useful

Nicole: Could use this to set up all the exemplary species.

3. Output in obo annotation format - or at least in form that can merge with. But - would still keep XML based data adapters. (DS noted that XML output needed for FB).

Nicole - likely to be more discussion of this new format in a couple of weeks.


web vs stand-alone

Nescent likes web because distributed curators.

BIRN - will integrate phenote into their apps.

Melissa: I think stand-alone is better becuase much more configurable - so much better to start with this to iron out wrinkles; easy to share data files; easy to manipulate configuration.