Talk:PMID Table Mock-up - Revision history

Siegele: /* Comments from Debby */ new section

2011-06-24T15:53:14Z

‎Comments from Debby: new section

Siegele: /* Comments from Marcus */ new section

2011-06-24T15:52:53Z

‎Comments from Marcus: new section

Siegele: Blanked the page

2011-06-24T15:52:06Z

Blanked the page

Siegele: /* Comments from Debby */ new section

2011-06-24T15:51:38Z

‎Comments from Debby: new section

Siegele: Created page with '==Comments from Marcus== From: Chibucos, Marcus [MChibucos@som.umaryland.edu] Sent: Monday, June 20, 2011 12:02 PM To: adrienne zweifel; Deborah Siegele; Jim Hu; Michelle Giglio;…'

2011-06-24T15:50:31Z

Created page with '==Comments from Marcus== From: Chibucos, Marcus [MChibucos@som.umaryland.edu] Sent: Monday, June 20, 2011 12:02 PM To: adrienne zweifel; Deborah Siegele; Jim Hu; Michelle Giglio;…'

New page

==Comments from Marcus==
From: Chibucos, Marcus [MChibucos@som.umaryland.edu]
Sent: Monday, June 20, 2011 12:02 PM
To: adrienne zweifel; Deborah Siegele; Jim Hu; Michelle Giglio; Peter Uetz
Subject: RE: Phenotype annotation table mock-up

Hi Adrienne and all,

I have a few suggestions for the table:

1. I think we should just have one table, the second one with the "Relative To" column. If it is a so-called "state phenotype annotation" that column can be left blank. I think this will allow us to group annotations better and is closer to what GO does (effectively, I think). To have two tables would be like GO having two annotation file formats.

2. The "Species" column:
I think this column should be renamed to "Taxon" or "Taxonomic Unit" because not all annotation subjects are at the species level; that is, we might want to capture strain variation, and species just sounds limiting.

A further consideration with the species column is this: there are multiple pieces of information in there, e.g. species (actually genus and species), strain, and NCBI taxon ID... How do we parse this information? Is there a systematic way to enter this? What if we want to populate some sort of annotation file with this table later? Might it be difficult to do this automatically with a script if we have multiple bits of information in a given field? I suggest that we have a separate column for each piece of information.

3. The "OMP ID" and "Term Name" columns:
I think that the "Term Name" column should be renamed "OMP Term Name" because it is more descriptive, and makes it clear that we're talking about the OMP term. This is important in order to not introduce confusion with the evidence term name (see below).

4. "Evidence" column:
"Evidence" is too generic, and doesn't describe what is to be entered in this field. I think we should have two columns for evidence, since we are planning to use ECO: "ECO ID" and "ECO Term Name". This will parallel the OMP information above.

My 2 cents.

← Older revision		Revision as of 15:53, 24 June 2011
Line 24:		Line 24:

	My 2 cents.		My 2 cents.
		+
		+	== Comments from Debby ==
		+
		+	*Change "Genus and species" to "Organism"
		+	*For E. coli, at least, need to add substrain
		+	*Come up with more user-friendly names for tables

← Older revision		Revision as of 15:52, 24 June 2011
Line 1:		Line 1:
−	~~==Comments from Marcus==~~
−	~~From: Chibucos, Marcus [MChibucos@som.umaryland.edu]~~
−	~~Sent: Monday, June 20, 2011 12:02 PM~~
−	~~To: adrienne zweifel; Deborah Siegele; Jim Hu; Michelle Giglio; Peter Uetz~~
−	~~Subject: RE: Phenotype annotation table mock-up~~

−	~~Hi Adrienne and all,~~
−
−	~~I have a few suggestions for the table:~~
−
−	1. I think we should just have one table, the second one with the "Relative To" column. If it is a so-called "state phenotype annotation" that column can be left blank. I think this will allow us to group annotations better and is closer to what GO does (effectively, I think). To have two tables would be like GO having two annotation file formats.
−
−	~~2. The "Species" column:~~
−	I think this column should be renamed to "Taxon" or "Taxonomic Unit" because not all annotation subjects are at the species level; that is, we might want to capture strain variation, and species just sounds limiting.
−
−	A further consideration with the species column is this: there are multiple pieces of information in there, e.g. species (actually genus and species), strain, and NCBI taxon ID... How do we parse this information? Is there a systematic way to enter this? What if we want to populate some sort of annotation file with this table later? Might it be difficult to do this automatically with a script if we have multiple bits of information in a given field? I suggest that we have a separate column for each piece of information.
−
−	~~3. The "OMP ID" and "Term Name" columns:~~
−	I think that the "Term Name" column should be renamed "OMP Term Name" because it is more descriptive, and makes it clear that we're talking about the OMP term. This is important in order to not introduce confusion with the evidence term name (see below).
−
−	~~4. "Evidence" column:~~
−	"Evidence" is too generic, and doesn't describe what is to be entered in this field. I think we should have two columns for evidence, since we are planning to use ECO: "ECO ID" and "ECO Term Name". This will parallel the OMP information above.
−
−	~~My 2 cents.~~
−
−	~~== Comments from Debby ==~~
−
−	*Change "Genus and species" to "Organism"
−	*For E. coli, at least, need to add substrain
−	*Come up with more user-friendly names for tables

← Older revision		Revision as of 15:51, 24 June 2011
Line 23:		Line 23:

	My 2 cents.		My 2 cents.
		+
		+	== Comments from Debby ==
		+
		+	*Change "Genus and species" to "Organism"
		+	*For E. coli, at least, need to add substrain
		+	*Come up with more user-friendly names for tables