Question

rice annotation package

0

Entering edit mode

zhang.jianhai ▴ 10

@zhangjianhai-12955

Last seen 4.3 years ago

Hello,

Is there an equivalent package for rice Affymetrix like Arabidopsis (https://bioconductor.org/packages/release/data/annotation/html/ath1121501.db.html)?

I want to retrieve the annotations by using rice affymetrix ids.

Regards,

Jianhai

rice annotation package • 1.5k views

ADD COMMENT • link updated 5.9 years ago by James W. MacDonald 65k • written 5.9 years ago by zhang.jianhai ▴ 10

0

Entering edit mode

Is this what you mean riceprobe ?

ADD REPLY • link 5.9 years ago shepherl 3.8k

score 2 · Accepted Answer · 2018-05-17

2

Entering edit mode

James W. MacDonald 65k

@james-w-macdonald-5106

Last seen 15 hours ago

United States

There isn't a rice annotation package, nor is there any facility to build one yourself. Options would be to

Download the annotation csv from the Affy website and roll yer own
Use biomaRt
Use whatever you get from GEO, which is pretty much what the submitter got from Affy

I would choose the latter, all things equal. Something like

> library(biomaRt)
> library(GEOquery)
> z <- getGEO("GSE4471")[[1]])
> mart <- useMart("plants_mart", "osativa_eg_gene",host = "plants.ensembl.org")
> annot <- getBM(c("affy_rice","description","ensembl_gene_id","entrezgene"), "affy_rice", featureNames(z), mart)

will give you the description, Ensembl ID and Entrez Gene ID. You can get other things as well. See the biomaRt vignette for more information.

Also do note that the results from biomaRt WON'T be in the same order as the query, so you need to ask to get the affy_rice IDs back, so you can use match to realign to your data. If you are using limma for analysis, you can then put that annot object into your ExpressionSet

eset <- rma(ReadAffy())

fData(eset) <- annot

And that will be carried through to your output topTable.

ADD COMMENT • link 5.9 years ago James W. MacDonald 65k

0

Entering edit mode

What I said about no facility to build one yourself isn't 100% correct. You could hypothetically build one using AnnotationForge's makeChipPackage along with an OrgDb from AnnotationHub:

> hub <- AnnotationHub()
snapshotDate(): 2018-04-23
> query(hub, c("oryza","orgdb"))
AnnotationHub with 9 records
# snapshotDate(): 2018-04-23
# $dataprovider: ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/
# $species: Aspergillus oryzae_RIB40, Bipolaris oryzae_ATCC_44560, Magnaport...
# $rdataclass: OrgDb
# additional mcols(): taxonomyid, genome, description,
#   coordinate_1_based, maintainer, rdatadateadded, preparerclass, tags,
#   rdatapath, sourceurl, sourcetype
# retrieve records with, e.g., 'object[["AH61938"]]'

            title                                               
  AH61938 | org.Oryza_sativa_(japonica_cultivar-group).eg.sqlite
  AH61939 | org.Oryza_sativa_Japonica_Group.eg.sqlite           
  AH61940 | org.Oryza_sativa_subsp._japonica.eg.sqlite          
  AH62214 | org.Oryza_brachyantha.eg.sqlite                     
  AH62580 | org.Magnaporthe_oryzae_70-15.eg.sqlite              
  AH62596 | org.Aspergillus_oryzae_RIB40.eg.sqlite              
  AH62653 | org.Bipolaris_oryzae_ATCC_44560.eg.sqlite           
  AH63011 | org.Trichosporon_oryzae.eg.sqlite                   
  AH63381 | org.Xanthomonas_oryzae_pv._oryzae_PXO99A.eg.sqlite

But those are 'bare' SQLite databases, and the functions in AnnotationForge really want an installed OrgDb package. So you would either have to create an actual OrgDb package, or change the code in AnnotationForge to accept a bare SQLite database.

ADD REPLY • link 5.9 years ago James W. MacDonald 65k

0

Entering edit mode

That's very useful. How can I get all the host names? It seems there is not listHosts() function.