deletion, insertion, insertion deletion

[noedelta]

On feature type.

We use features to describe the state of interactor in the interaction. For constructs which have these and are part of the interaction, without any immediate effect on the interaction, but possibly role in constitutive expression we cannot currently enter using any term. insertion deletions even when they have an effect on interaction cannot be covered right now. Could we have a heirarchy to describe insertion, deletion and insertion deletions the wy we can do mutations?

Reported by: jyotikhadake

[noedelta]

Logged In: YES user_id=653048

It is a problem to add the various types of mutation due to the existing children of mutation: mutation increasing/ decreasing interaction. If we add deletion, insertion, substitution we than have a little graph explosion, where each of these new term should have 2 children increasing and decreasing interaction.

Luisa

Original comment by: luisa_montecchi

[noedelta]

Logged In: YES user_id=1107300

Mutations are currently defined as upto 2 AA. I am talking of more than 2 amino acid so these cannot be classified as mutations.

Original comment by: jyotikhadake

[noedelta]

Logged In: YES user_id=653048

An easy solution (not ontologically correct) is to add 2 new terms in feature detection method similar to 'deletion analysis' : 'insertion analysis' and 'substitution analysis'

All could be children on 'mutation analysis'.

Luisa

Original comment by: luisa_montecchi

[noedelta]

Logged In: YES user_id=719654

per Luisa's request - here are additional comments brought from an external email thread that seems to be relevant to the terms in question:

Andrew Chatr-aryamontri wrote:

> Dear all, > the {IMEx - LS] curation manual at page 16 states: > > A mutation (hotspot) may be defined by deletion analysis > but only by the deletion of a single or two adjacent > amino acid residues. If a deletion is 3 amino acids or > greater it should be annotated as <featuretype> "binding > site" <featureIdentification> "deletion analysis" not a > <featureType> "mutation" > > We already discussed about this point in Los Angeles > and was suggested to write down a list of exceptions > to the rule. Unfortunately it is extremely difficult > to produce this list since the rule by itself is an > exception. The adoption of this rule results in a > misinterpretation and misrepresentation of experimental > results. > Experimentally a binding range is an aminoacidic range > that was proved to bind a partner. The deletion of a > sequence by itself doesn’t provide this information. > It only proves its relevance in the interaction and > this information is well represented by the “mutation” > term. Furthermore short deletions display their effect > by disrupting domains involved in binding but the > deleted sequence by itself doesn’t have any relevant > binding property. There isn’t any rationale (apart that > we had already agreed on it) supporting the manual > statement. > > We would suggest to change the rule in: > …..If a deletion is 3 amino acids or greater it CAN > be annotated as <featuretype> "binding site"……

with me following-up:

> I obviously haven't spent too much time thinking > about this but maybe it would make sense to somehow > separate what was done (deletion, mutation) from > the actual effect the change makes and/or conclusions > regarding the binding sites/regions ?

lukasz

Original comment by: lukasz99

[noedelta]

Logged In: YES user_id=653048

Approved at the PSI Washington meeting and updated fetaure detection methods with the following terms. mutation analysis -deletion analysis -insertion analysis -substitution analysis --alanine scanning

Original comment by: luisa_montecchi

[noedelta]

• status: open --> closed

Original comment by: luisa_montecchi