MI:0808 (comigration in sds page) clarification/fixes

[lukasz99]

Hello,
can someone confirm/deny that this term, as currently defined:

Method allowing the detection of strong interactions between two
molecules by their very close proximity or the overlap of their
relative bands in a denaturing SDS gel.

applies only to hetero-dimers ?

If it, indeed, does, it should be modified as 'their very close proximity' seems to
refer to any two molecules (not bands) which, in turn, seems to include
homo-dimers - note that the band corresponding to a protein that runs at
twice (three, four, etc times) its molecule weight implies 2,3,4, etc molecules
are running together so, technically, it is covered by the current definition.

If the terms is supposed to cover a more generic case of any complex
it should clearly say so without explicitly referring to exactly two
molecules.

if it is somehow advantageous to separate homo-oligomer detection from
hetero-oligomer detection the definition should be modified to clearly say
so. Note that it would imply homo-oligomers could be annotated using
only parent term unless a mixture of differentially labeled (or tagged)
molecules is used and each is detected separately. Fair and square,
but it this is the case it should be clearly stated in ht definition.

Comment: Personally, I don't think there's much to be gained by treating
homo-oligomers in some special way - the experiments show
that the interacting molecules (2, 3 or more) comigrate in sds-page

• this is independent of their identity (participant identification annotation
  clarifies if it is because of the band position or by testing for participant
  presence). I would also substitute SDS with 'denaturing' gel - SDS seems to be
  the most popular for proteins but every now an then other denaturing agents
  (ie urea, particularly for nucleic acids) are used. Please, also note that
  running in close proximity (as stated in the definition) is not sufficient,
  unless all molecules run within a single band (and even then it is necessary
  but not sufficient condition - vide two proteins of the same molecular weight)

Also note that, in case homo-oligomers should be treated separately
from hetero-oligomers there are more terms to be split the same way
(eg. gel filtration & centrifugation based methods). These methods are
prone to the same problems as denaturing electrophoresis (every now
and then things do run at the wrong place - membrane proteins, proteins
of strange shapes and charge) so for consistency all of them should
be treated the same way.

My preferred version of the definition would read, more or less:

A method allowing the detection of strong interactions between two
or more molecules as running, all of them, within a single band in
a denaturing gel.

lukasz

[pporrasebi]

Agreed and definition amended to your proposal.

[pporrasebi]

On 6/4/2020 12:29 PM, Lukasz Salwinski wrote:

sigh... the modified definition now reads:

The interaction of two molecules is determine by their very close proximity or the overlap of their relative bands in a gel.

oh... and 'relative' bands look weird - the proper word would
be 'respective'

please also note that, whatever the changes are, they should
be propagated to the child terms.

I'm sorry for what might look like nit-picking but definitions should
be clear and unambiguous without asking the wise ones for interpretation.
lukasz

apart from missing 'd' in 'determine', it is still not clear how
to deal with homo-oligomers - I read the definition and homo-oligomers
seem ok but others might claim they are not. if only hetero-oligomers
are ok then the definition should mention 'distinct molecules'.

to make the definition worse, it now says 'two molecules' - does it
mean the term cannot be applied to heterotrimers ??? I've got no
problem imagining three (or more) subunits of a membrane complex
running together even on sds-page gel :o(

lukasz

[pporrasebi]

Definition amended, here's the new one:

"The interaction of two or more molecules is determined by their very close proximity or the overlap of their respective bands in a gel."

Pushing it to the repo in a few minutes, please suggests any changes if you see fit.

[pporrasebi]

By the way, related terms have also been amended accordingly.