search

HUPO-PSI / psi-mi-CV

Molecular Interactions Controll Vocabulary
Creative Commons Attribution 4.0 International
4 stars 4 forks source link

proteomic level studies and interaction type #59

Closed noedelta closed 8 years ago

noedelta commented 18 years ago

Proteomic scale studies can be entered as interactions for example cases detailed below. It is however not quite right to add these using physical interactions as interaction type. An alternative interaction type is needed for these cases: PMID: 12577067 ICAT or SILAC methods used to determine the increase or decrease in proteins in response to ligands. Purification based on antibodies to a domain of a protein. PMID: 16236029 Postsynaptic densities (vesicles) purified after many rounds of density gradient purifications and proteins in these identified. In thsi case colocalisation will be as appropriate or inappropriate as physical interaction.

Reported by: jyotikhadake

noedelta commented 18 years ago

Logged In: YES user_id=653048

Icat and I guess Silac are quantification method not interaction method and currently they are not easy to assign to some existing CV. Reading the abstract of 12577067 I deduce that the interaction are detected by affinity purification. Would you create interaction with EGF for all the other protein 'responsive' to EGF. A possible interaction type could be 'quantitive response'. But it is at the limit of PSI-MI scopes.

For the second case why not use cosedimentation through density gradients MI:0029 as interaction method, mass spec as participant detection and 'co-localisation' interaction type.

Original comment by: luisa_montecchi

noedelta commented 18 years ago

Logged In: YES user_id=653048

first case: For the SILAC paper these are physical interaction with some extra quantitative/functional info. The method is a combined pull down + SILAC , but as we do not store quantitative measure nor have a new CV type 'quantitative measurement' the exp-modification is fair enough for now. This discussion should be also forwarded to PSI/IMEX partners.

second case : interaction type for the post-synaptic proteome (collins et al). Although there are many purification steps involved there is still membrane around and no evidence there is only one type of complex. A new term for interaction type could be 'subcellular component' as child of co-localisation. This issue should be discussed at PSI level (and IMEX as these data may not be exchanged) so for the time being 'co-localisation' is the best interaction type.

Original comment by: luisa_montecchi

noedelta commented 18 years ago

Original comment by: luisa_montecchi