inconsistent use of "cytoplasmic"

[gocentral]

It seems that the majority of uses of the adjective "cytoplasmic" in the GO actually refer to GO:cytosol and not GO:cytoplasm

GO:0005737 cytoplasm [DEF: "All of the contents of a cell excluding the plasma membrane and nucleus, but including other subcellular structures."]

GO:0005829 cytosol [DEF: "That part of the cytoplasm that does not contain membranous or particulate subcellular components."]

The mitochondrion (for example) is part_of the cytoplasm, both according to the definition of cytoplasm, and according to the graph.

However, it seems that "cytoplasmic X" and "mitochondrial X" terms are is_a siblings, rather than is_a parent-child, as one would expect if the adjectival qualifier is refering to partonomy or location.

For example:

GO:0016482 cytoplasmic transport [DEF: "The directed movement of substances (such as macromolecules, small molecules, ions) into, out of, or within the cytoplasm of a cell."]

Is an is_a sibling of:

GO:0006839 mitochondrial transport [DEF: "Transport of substances into, out of or within a mitochondrion."]

However, within a mitochondrion necessarily means within a cytoplasm, so mt transport is_a ct transport.

This was discussed for ER-ubiquitib ligase complex and chromosome in an email exchange

Re: what does "found in the cytoplasm" mean? May 12

There was no firm conclusion, but it seemed the preference was to (a) keep the wording "cytoplasmic" and (b) understand the meaning as being "cytosolic"

I favour a more extreme solution of making this absolutely crystal clear and renaming all terms as "cytosolic", unless the meaning is genuinely "cytoplasmic". The definitions would be changed as well.

This is probably unlikely so I would at least like to see synonyms added (possibly with a FORMAL synonymtypedef) and the definitions changed to make it absolutely clear.

Reported by: cmungall

Original Ticket: "geneontology/ontology-requests/5077":https://sourceforge.net/p/geneontology/ontology-requests/5077

[gocentral]

Logged In: YES user_id=254723 Originator: YES

see also: http://wiki.geneontology.org/index.php/XP:cellular\_component\_xp\_self

Original comment by: cmungall

[gocentral]

Logged In: YES user_id=1101528 Originator: NO

A slightly tangential comment - We've tripped over this in Reactome annotation exactly as Chris describes and have evaded the problem for ourselves by banning the use of "cytoplasmic" to describe the location of a physical entity. Initially, I thought this would cause major problems, expecting to frequently encounter published evidence that was good enough to locate a physical entity somewhere between the cell membrane and the nucleus, but no more. In practice, that problem has not arisen, and the number of cases where a curator has had to infer a location more precise than the one stated by the authors of a paper is small.

So there's a dilemma - synonyms make the ambiguity explicit but do not resolve it (and the people we talk to about reasoning over location data insist that this ambiguity is fatal), but cleaning up the legacy ambiguities looks like an impossibly large chore. Unless others, like us, discover that "cytoplasmic" essentially always was used to mean "cytosolic" in which case the 'extreme solution' works.

Peter

Original comment by: deustp01

[gocentral]

Logged In: YES user_id=436423 Originator: NO

In GO, I think we'll have to address these on a case-by-case basis; I hope it won't turn out to be an impossibly large chore. For macromolecular complexes, "cytosolic" is probably the intended meaning, so we can change names, but for the transport example above I'd be more inclined to keep the "cytoplasmic" name and move the mitochondrial transport term (unless the transport wg documents a case for renaming the cytoplasmic transport term instead).

Chris - is there a report of these somewhere?

> There was no firm conclusion, but it seemed the preference was to (a) keep > the wording "cytoplasmic" and (b) understand the meaning as being > "cytosolic"

I actually don't like this idea at all. It got suggested towards the end of the email thread, and I didn't have time to comment on it before I went off on holiday. I hereby emphatically vote for names to match meanings.

m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=554670 Originator: NO

Cytosol is often used to refer only to the soluble fraction of the cytoplasm. You won't know this until you've done the experiment - fractionate away organelles and membranes and then spin longer at a higher speed to spin out the large protein complexes and insoluble bits. What's left is the cytosol.

You localize a protein via immunofluorescence. It shows up as punctate spots and is not colocalized with any organelle or membrane system. I would feel uncomfortable annotating this to cytosol,

My $0.02 - it's more conservative to keep things "cytoplasmic" unless you know for sure its "cytosolic".

Original comment by: eurie

[gocentral]

Original comment by: cmungall

[gocentral]

Logged In: YES user_id=254723 Originator: YES

Eurie:

would this structure work?

cytoplasmic X --[i] cytosolic X --[i] mitochondrial X nuclear X

This is logically consistent, and allows for varying degrees of knowledge in annotation.

Of course, getting back to Midori's point, we would probably have to go through on a term-by-term (hopefully not annotation-by-annotation) basis and see if existing terms "cytoplasmic X" really correspond to "cytoplasmic X" or "cytosolic X".

Midori:

I'll attach a report of all cytoplasmic terms, but really this is a just a text search like you could do in OE. I will attach the xp defs for the terms, if available, if that helps. We can also look at the reasoner results and see where the reasoner thinks the graph should be adjusted (e.g. the mitochondrial X under cytoplasmic X cases)

File Added: go-cytoplasmic.txt

Original comment by: cmungall

[gocentral]

Logged In: YES user_id=1101528 Originator: NO

Chris,

Wouldn't your structure

cytoplasmic X --[i] cytosolic X --[i] mitochondrial X nuclear X

need to be extended to have a separate --[i] vesicular X for every cytoplasmic subcompartment (vesicular here is a stand-in), a list that runs to at least dozens in mammals (and probably nearly as many in yeasts, and a partly disjunct large set in plants)?

Original comment by: deustp01

[gocentral]

Logged In: YES user_id=554670 Originator: NO

I guess theoretically, having a "cytoplasmic X" and "cytosolic X" would work but it seems like you wouldn't want to create the cytosolic term unless you had evidence for this, which may be an annotation by annotation process.

Original comment by: eurie

[gocentral]

Logged In: YES user_id=436423 Originator: NO

Chris and I looked at the 'cytoplasmic' terms again and decided that no further changes are needed right now. We also agreed -- and should document -- that we'll carry on using 'cytoplasmic' and 'cytosolic' in term names so that they match the meanings and definitions. For many of the existing 'cytoplasmic' terms it's possible, even likely, that the actual entity is cytosolic, but we have no organelle terms to accommodate, nor do we necessarily have enough experimental evidence to be sure (essentially Eurie's last comment).

Original comment by: mah11

[gocentral]

• labels: --> Other term-related request
• milestone: --> TermGenie
• assigned_to: nobody --> gomidori
• status: open --> closed-fixed

Original comment by: mah11