Specifications for a DAF for gene-disease data

New DAF specs as specified by the AGR Disease Working Group

• These specifications have been discussed and specified by the Disease Working Group of the AGR

https://docs.google.com/spreadsheets/d/1PrUI8CwV7AejBloWG2e6SNXEz_FSpCeMEwHhcKE9KmU/edit?usp=sharing and scroll to the tab 'DAF spec-v2 (for developers)'

• Annotations from the computational pipeline that WB currently has for assigning 'Potential models' will not be included in the MOD-specific DAFs. The plan is to have a separate AGR orthology pipeline to link human genes and their orthologs from model organisms.

• Header for file:

!daf-version: 0.1
!Date: 02/08/2017
!Project_name: WormBase (WB) Version WS258
!URL: http://www.wormbase.org/
!Contact Email: wormbase-help@wormbase.org
!Funding: NHGRI at US NIH, grant number U41 HG002223

• Use column headers, from 'Content' column on tab 'DAF spec-v2 (for developers)' here:

https://docs.google.com/spreadsheets/d/1PrUI8CwV7AejBloWG2e6SNXEz_FSpCeMEwHhcKE9KmU/edit?usp=sharing

• Name of file: disease_association.WB.WS258.txt
• Source of DO ID translation to name should be from the DO file:

raw.githubusercontent.com/DiseaseOntology/HumanDiseaseOntology/master/src/ontology/doid-merged.obo

Changes to DAF

Changes official as of the DWG-DQM call, July 5th, 2017

• Column 3: DB: which was the database from which the identifiers in 'DB object ID' and 'DB Object Symbol' are drawn, is now deleted (Column 3 is now DB Object ID)
• Column 4: DB Object Symbol: Doesn't need the database prefix 'WB:" as it's tied to the ObjectID field which does
• Column 7: is new, for additional genetic components, note that DQM's script will fill in the symbol for the required format: <DB_object_id>;<DB_object_symbol>
• Column 9: Association Type has only 3 allowed values, these values are no longer valid: causes_or_contributes _to_condition, causes_condition, contributes_to_condition
• Column 12: With Ortholog, mostly for SGD

Rules for DAF/JSON

1. Selecting DB Object:
   • In the majority of our annotations Disease_relevant_gene is going to be present, pick this (Gene) as the DB Object for DAF/JSON if Strain, Variation, Transgene are all absent.
   • If either Strain, Variation or Transgene is present, pick that as the DB Object for DAF, even though Disease_relevant gene is present. If more than one of these is present, then the order should be Strain>Variation>Transgene.
   • Note that Inferred Gene will most likely be present for a majority of our annotations, it's value which is a gene will be redundant with the Disease relevant gene. This is okay and required for AGR.
   • Will clean up the data so that only one of Strain, Variation or Transgene is present.
2. For WS262:
   • will have to hack this for DAF/JSON: Change Association type: causes_condition, causes_or_contributes_to_condition, and contributes_to_condition to is_implicated_in (all three have been deprecated by AGR DWG), this will be changed in the data for WS262.

DAF columns: definitions and content

DAF columns, allowed values and mapping of columns from old WB DAF

Column	Content	Required(R) Optional(O)	Cardinality	Example	Definition	Old DAF column	Acedb tag	Required format/Restriction
1	Taxon	R	1	taxon:6239	NCBI taxonomic identifier for the organism	13	No acedb tag	taxon:####
2	DB Object Type	R	1	gene, allele, transgene, genotype, fish	The type of object being annotated	2	Modeled_by	WB uses gene, alllele, strain or transgene
3	DB	R	1	WB	The database from which the identifiers in 'DB object ID' and 'DB Object Symbol' are drawn		No Acedb tag	WB
4	DB Object ID	R	WBGene00004887	A unique identifier from the database in DB for the entity being annotated; note, no DB prefix	2	Modeled_by	DB:<ID>
5	DB Object Symbol	R	1	smn-1	A (unique and valid) symbol to which DB object ID is matched	3	Modeled_by	<symbol>
6	Inferred Gene Association	O	0 or greater	WB:WBGene00004887	Database ID for inferred gene/marker association that can be made based on the DB object ID	Does not exist	Inferred_gene	WBGene ID; repeat WBGene ID here even if 'DB Object Type' is gene
7	Gene Product Form ID	O	0 or 1	UniProtKB id or PRO ID	this field allows the annotation of specific variants of that gene or gene product	Does not exist	No acedb tag	Blank, WB not using this for now
8	Experimental Conditions (to create the model)	O	0 or greater	standard conditions chemical/drug treatments (ChEBI ID, ZECO) dietary manipulations (specifiy entity) surgery/amputation bacterial/virus exposure (taxon ID)	Experimental/environmental (i.e. non-genetic) conditions required for the model, used particularly for induced models	Does not exist	Inducing_chemical, Inducing_agent	WB is using free text for now, will move to using an ontology soon
9	Association Type	R	1	is_model_of is_implicatd_in is_marker_for	Relationship between the DB object and the disease	does not exist	Association_type	is_model_of' only for DB object type: 'genotype, strain, organism, Fish' is_implicated in only for alleles, genes, transgenes model change did not make into WS261, so for now replace: causes_or_contributes _to_condition, causes_condition, contributes_to_condition with 'is_implicated_in'
10	Qualifier	O	0 or 1	NOT	Used to indicate that the DB object is not associated with the DO term/association type	Does not exist	Qualifier	NOT
11	DO ID	R	1	DOID:12858	DO identifier for disease	5	Disease_term
12	With	O	0 or greater	DB:gene_symbol DB:gene_id DB:gene_symbol[allele_symbol] DB:allele_id	EITHER specifies additional genetic components of the model (where DB Object Type is not genotype/strain/fish)OR specifies the orthologous (usually human) gene in annotations with ‘ISS/ISO’ evidence code	Does not exist	No acedb tag	Blank, mostly for SGD, mandatory for ISS/ISO evidence codes
13	Modifier Association Type	O	0 or 1	condition_ameliorated_by condition_exacerbated_by	Relationship between the modifier and the disease model	Does not exist	condition_ameliorated_by condition_exacerbated_by	WB has started using
14	Modifier - qualifier	O	0 or 1	NOT	Used to indicate DB object is not associated with the DO term/association type	Does not exist	Need to add tag for WS262	WB has started using
15	Modifier (genetic)	O	0 or greater	DB:gene_symbol DB:gene_symbol(allele_symbol) DB:gene_id DB:allele_id	Specifies a genetic object (allele or gene) that modifies the disease model	Does not exist	Modifier_transgene Modifier_variation Modifier_strain Modifier_gene	WB will start using
16	Modifier - experimental conditions	O	0 or greater	standard conditions chemical/drug treatments (ChEBI ID, ZECO) dietary manipulations (specifiy entity) surgery/amputation bacterial/virus exposure (taxon ID)	Specifies a non-genetic object experimental condition that modifies the disease model	Does not exist	Modifier_molecule Other_modifier	WB will start using
17	Evidence Code	R	1 or greater	EXP, IMP, IPM, IGI, IDA, IED, IEP, IAGP, ISS, ISO, TAS, IC, IEA	From GO: Indicates the kind of evidence in the cited source that supports the disease annotation. If reference describes multiple methods that each provide evidence, then multiple annotations should be made with same DO term and different evidence codes	7	Evidence_code	'IMP' for now, other evidence codes coming soon
18	Genetic sex	O	0 or 1	male/female/hermaphrodite	Genetic sex of the model	Does not exist	Genetic_sex	Use value 'Hermaphrodite' for all rows May use 'male' or 'female' in the future if authors specify
19	DB:Reference	R	1	PMID:14978262	unique identifier(s) for a single source cited as an authority for the attribution of the DO ID to the DB object ID	6	Paper_evidence	Use only PMIDs for AGR 0.3, will get to individual DB identifiers later on
20	Date	R	1	20090118	Date on which the annotation was made; format is YYYYMMDD	14	Date_last_updated
21	Assigned By	R	1	WB	The database which made the annotation - one of the values from the set of GO database cross-references	15	No acedb tag

Old pre-AGR WormBase DAF

DAF 2.0 for gene-disease data includes all genes with the Experimental_model and/or Potential_model tags.

Format: The gene-disease association file is a 17 column tab-delimited file, where 11 columns have to have data and 6 are optional.

Mapping of GAF column to gene-disease data

Column	Column Name	Required?	Cardinality	Example
1	DB	required	1	WB
2	DB Object ID	required	1	WBGene00007799
3	DB Object Symbol	required	1	nrx-1
4	Qualifier	optional	0 or greater	(no value, leave empty)
5	GO ID	required	1	DOID:0060041
6	DB:Reference (|DB:Reference)	required	1 or greater (separate values with pipes)	WBPaper00041363|WBPaperXXXXXXXX
7	Evidence code	required	1	IMP or ISS (use 'IMP' for Experimental_model genes and ISS for Potential_model genes)
8	With (or) From	optional	0 or greater (separate values with pipes)	OMIM:600565|OMIM:600566
9	Aspect	required	1	D (D for disease ontology, all annotations have this value)
10	DB Object Name	optional	0 or 1	(no value, leave empty)
11	DB Object Synonym (|Synonym)	optional	0 or greater	(no value, leave empty)
12	DB Object Type	required	1	gene (all annotations have this value)
13	Taxon	required	1 or 2	taxon:6239
14	Date	required	1	20130422 (this is the date of annotation dumped under Date_last_updated; for potential_model genes, use the date on which the OMIM homology script is run)
15	Assigned By	required	1	WB
16	Annotation Extension	optional	0 or greater	(no value, leave empty)
17	Gene Product Form ID	optional	0 or greater	(no value, leave empty)

Back To Disease and Drugs