IMP

From GO Wiki
Jump to navigation Jump to search

See IMP in evidence docs.

This page is for additional comments on this code

WITH column and IMP

The docs state:

We recommend making a "with" entry in the with/from column when using this evidence code to indicate the identifier for the allele in which the phenotype was observed. When multiple entries are placed in the with/from field, they are separated by pipes.

Example for how the with/from column should be filled in

  • The mouse gene product Actc1 (actin, alpha, cardiac ; MGI:87905), has a GO annotation to muscle thin filament assembly ; GO:0030240, inferred from mutant phenotype, IMP of MGI:2180072 (symbol: Actc1tm1Jll; name: targeted mutation 1, James Lessard), from PMID:9114002 external link. MGI:2180072 is entered in the with/from column for this annotation.


ZFIN

ZFIN uses a mixture of

  • genotype IDs (e.g. ZFIN:ZDB-GENO-000412-4)
  • morpholino IDs (e.g. ZFIN:ZDB-MRPHLNO-071219-3)

WB

WB uses

  • worm phenotype ontology IDs e.g. WB:WBPhenotype0001020
  • RNAi experiment IDs e.g. WB:WBRNAi00000010

Note that worm uses prefixed pheno IDs (WB:WBPhenotype0001020). However, the WP ontology uses unprefixed IDs (WBPhenotype0001020) - this causes hiccups in data integration

MGI

MGI uses

  • allele IDs e.g. MGI:MGI:3762127

Note that the MGI make available downloads for genotype-phenotype associations - however, these reports use genotype symbols, and not allele IDs. allele->MP mappings available from MGI on request

Example:

Ephb2tm1Paw allele has various phenotypes - in mh2 and cx5 genotypes it displays "abnormal optic tract morphology"

If we look at the GO annotations for which an IMP is used as evidence with this allele:

MGI MGI:99611 Ephb2 GO:0007612 ! learning PMID:11754835 IMP MGI:MGI:2149765 P Eph receptor B2 Drt|Erk|eteck|Hek5|Nuk|Prkm5|Qek5|Sek3|Tyro5 gene taxon:10090 20031124 MGI
MGI MGI:99611 Ephb2 GO:0048168 ! regulation of neuronal synaptic plasticity PMID:11754835 IMP MGI:MGI:2149765 P Eph receptor B2 Drt|Erk|eteck|Hek5|Nuk|Prkm5|Qek5|Sek3|Tyro5 gene taxon:10090 20031124 MGI
MGI MGI:99611 Ephb2 GO:0048170 ! positive regulation of long-term neuronal synaptic plasticity PMID:11754836 IMP MGI:MGI:2149765 P Eph receptor B2 Drt|Erk|eteck|Hek5|Nuk|Prkm5|Qek5|Sek3|Tyro5 gene taxon:10090 20040414 MGI

The MGI structured notes column includes "text:abnormal optic tract morphology" (but no MP:0004267 ID) for the annotation to:

  • GO:0021631 ! optic nerve morphogenesis
  • GO:0048593 ! camera-type eye morphogenesis

Why the discrepancy?

RGD

RGD has a single entry

  • RGD:1322442

FB

FlyBase leaves the WITH column blank for IMPs

SGD

Blank

DictyBase

Blank

Pombe

  • GO IDs (e.g. for GO:0000145)
  • Pombe IDs (e.g. GeneDB_Spombe:SPAC17G6.12) all genes?
  • SGD IDs (e.g. SGD:S000006349) gene?
  • PMIDs (PMID:11679064)

Should there be PMIDs there?

GO IDs are unusual. Should these be ICs?

 GeneDB_Spombe   SPCC970.09      sec8            GO:0006887      PMID:11854409   IMP     GO:0000145      P       exocyst complex subunit Sec8    SPCC970.09|sec8 gene    taxon:4896      20041107        GeneDB_Spombe

e.g. GO:0006887=exoytosis, GO:0000145=exocyst. PMID:11854409 refers to a paper "The multiprotein exocyst complex is essential for cell separation in Schizosaccharomyces pombe". The mutant is sec8-1. Maybe this should be two annots, one IMP and one IC? Or two IMPs with the same paper? NOTE: Val has been alterted & this is being fixed

Should we allow allele symbols in the WITH column where there are no IDs (still may be hard for MODs to track in their internal dbs)

TIGR

Protein and PFAM IDs

GOA Human

3 entries only:

  • UniProtKB:O94979
  • UniProtKB:P49917
  • UniProtKB:Q8WUM4

TAIR

  • gene IDs (e.g. TAIR:GENE:3356228)

Appears to be used in cases of expression of another gene in a mutant background of another gene; see for example annotations for CPL1

TAIR do track polymorphism IDs, but they aren't used in the with column

  • ID prefixes are inconsistent w.r.t case (gene:nn in col2, Gene:nn on TAIR webpage, GENE:nn in with). Seemingly trivial but an obstruction to data integration
  • TAIR seem to record additional information re: what the with column means (e.g. background of another gene). Should we have a mechanism for capturing this in GO?


Gramene Oryza

Blank

Recommendations for standardizing the WITH column

  • something is better than nothing
  • not all MODs have IDs for alleles, genotypes or phenotype classes
  • geno/allele ID vs pheno ID

What sort of thing can be inferred from phenotype?

  • Regulation of X vs X?

WB have a mapping they use for RNAi experiments. Others are down to curator judgement