include additional values to CausalMechanismQualifierEnum

[vdancik]

Is your feature request related to a problem? Please describe.
We are missing a few mappings between ChEMBL's action types and CausalMechanismQualifierEnum:

Causal Mechanism Qualifier Enum	ChEMBL	ChEMBL definition
chelation	CHELATING AGENT	Binds to a metal, reducing its availability for further interactions
crosslinking	CROSS-LINKING AGENT	Induces cross-linking of proteins or nucleic acids
oxidation	OXIDATIVE ENZYME	Oxidises a substrate through enzymatic reaction
sequestering	SEQUESTERING AGENT	Binds to a substance such as a drug, toxin or metabolite reducing its availability for further interactions
exogenous_protein	EXOGENOUS PROTEIN	Protein from an exogenous source acts as a substitute or supplement for a specific protein which is absent or has reduced function in affected patients
exogenous_gene	EXOGENOUS GENE	Nucleic acid from an exogenous source acts as a substitute or supplement for a specific gene which is absent or has reduced function in affected patients

Describe the solution you'd like
add additional entries to CausalMechanismQualifierEnum

What working group (or team) did this request originate from?
MolePro

Note: This is relevant for members of NCATS Translator.

Additional context
needed for ChEMBL ingest

Tag relevant members for discussion
@sierra-moxon @mbrush

[mbrush]

Proposed updates to CausalMechanismEnum - to accommodate SIGNOR terms, and new Chembl terms above. Terms in this enum are informed mainly by SIGNOR (does not give definitions) and Chembl (definitions are here)

Note that Vlado suggests no longer making binding a top level parent of many of the mechanisms that require binding to happen - as this introduces an additional classification axis that makes organizing other terms difficult. This may be OK because of the convention that when a mechanism implies binding, we create an explicit 'directly_physically_interacts_with' edge. These binding-based mechanisms should be documented somewhere however (@vdancik @sierra-moxon)

So, the updated proposal is below. I guessed on a few of these - @vdancik please give a close look and move things around where warranted (you can copy your updated suggestion into a new comment).

* = new term for SIGNOR
** = new term for Chembl
*** = new term for GtoPdb
**** = new term for DGIdb
***** = new term for Intact

modulation**
	allosteric_modulation*****
	mixed_agonism***
	positive_modulation**
		potentiation
		induction
		cofactor****
		activation
			agonism
				partial_agonism**
				biased_agonism***
				non_competitive_agonism***
			positive_allosteric_modulation
			molecular_channel_opening
			stimulation
			guanyl_nucleotide_exchange*
	negative_modulation**
		negative_gene_editing_modulation**
		gtpase_activation*
		atpase_activation*****
		antisense_oligonucleotide_inhibition**
		rna_interference_inhibition****
		feedback_inhibition***
		suppression**
		inhibition
			antibody_inhibition
			antagonism
				allosteric_antagonism**
			competitive_inhibition***
			noncompetitive_inhibition
				negative_allosteric_modulation
			gating_inhibition***
			irreversible_inhibition***
			molecular_channel_blockage
			inverse_agonism
	binding
	covalent_binding 
		adduction****
		crosslinking
		transglutamination 
		disulphide_binding
	stabilization
		chaperone_mediated_stabilization**
	destabilization*?
	degradation**
	cleavage**
	hydrolysis**
	disruption**
	opening*
	multi-target_modulation****

	chelation**
	release**
	reduction**
	sequesteration**
	oxidoreduction*****
	oxidation**
	exogenous_gene**
	exogenous_protein**
	transcriptional_regulation
	translational_regulation*
	catalytic_activity*
	chemical modification* (modification of a chemical entity, by a gene product)
	relocalization*
	isomerization* (target undergoes isomerization as a result of association with effector)
	signaling_mediated_control
	immune_system_modulation****
		vaccine_antigen**
	post-transcriptional regulation*
	molecular modification	/ post-translational modification (addition of moieties)
		phosphorylation*
		dephosphorylation*
		neddylation*
		deneddylation*
		lipidation*
			palmitoylation*
			myristoylation*****
		tyrosination*
		carboxylation*
		ubiquitination*
			monoubiquitination*
			polyubiquitination*
		deubiquitination*
		sulfation
		acetylation*
		oxidation*
		acetylation*****
		deacetylation*
		glycosylation*
		deglycosylation*
		methylation*
			trimethylation*
		demethylation*
		sumoylation*
		desumoylation*
		adp_ribosylation*
		de_adp-ribosylation*****
		ampyulation*****
		hydroxylation*
		s-nitrosylation*

[mbrush]

Proposed mapping of SIGNOR 'mechanisms' to values in this enum: https://docs.google.com/spreadsheets/d/1tqimhXxpWzQdfNxanpW-rAaLnmsP80YZUthY5O4mEc8/edit?gid=1199093337#gid=1199093337&range=A1

Proposed mapping of GtoPdb'mechanisms' to values in this enum: https://docs.google.com/spreadsheets/d/1tqimhXxpWzQdfNxanpW-rAaLnmsP80YZUthY5O4mEc8/edit?gid=1199093337#gid=1199093337&range=A47

TO DO: create mapping to DGIdb mechanism terms here: https://dgidb.org/about/overview/types-and-directionality. Colleen's mappings for DGIdb are here.

TO DO : Create mappign to Chembl terms

TO DO: update enumeration hierarchy in Biolink, and add/include mappings to terms from the sources above (and others)

[mbrush]

@vdancik please provide examples of exogenous_gene and exogenous_protein as mechanisms - so we can define them correctly.

[vdancik]

@vdancik please provide examples of exogenous_gene and exogenous_protein as mechanisms - so we can define them correctly.

PROTHROMBIN COMPLEX CONCENTRATE
VORETIGENE NEPARVOVEC

[mbrush]

Thanks Valdo - in all these examples the subject/effector is just an exogenous version of the target gene/protein. Not really valuable edges in my opinion. I say we filter out records with these mechanisms/action _types.