[ensembl-dev] REST eQTL questions

[Thomas Juettemann]

Hello Andrew,

my apologies for the delay in getting back to you and thank you for your interest in the eQTL data!
Please find the answers inline, and don't hesitate to contact us again with question you may have, they are certainly not naive.

Best wishes,
Thomas


On 25 Jan 2017, at 02:26, andrew126 at mac.com wrote:

Hi,

Please forgive some naive REST eQTL questions.

The documentation says:

GET eqtl/stable_id/:species/:stable_id	Returns the p-value for each SNP in a given gene (e.g. ENSG00000227232)
1)
How is “in a given gene” defined?

If I query REST for the homo sapiens gene in the example/documentation (ENSG00000227232) without any tissue or variant-name restrictions, then 159,122 of the returned results have a “display_consequence” of “intergenic_variant”, which is unexpected relative to “in a given gene”.

The counts for each different display_consequence are shown below:

   6366     'display_consequence' => '3_prime_UTR_variant'
   1070     'display_consequence' => '5_prime_UTR_variant'
  21006     'display_consequence' => 'downstream_gene_variant'
      2     'display_consequence' => 'inframe_deletion'
 159122     'display_consequence' => 'intergenic_variant'
 186614     'display_consequence' => 'intron_variant'
   1078     'display_consequence' => 'missense_variant'
   6794     'display_consequence' => 'non_coding_transcript_exon_variant'
     86     'display_consequence' => 'non_coding_transcript_variant'
    634     'display_consequence' => 'splice_region_variant'
    948     'display_consequence' => 'synonymous_variant'
  21868     'display_consequence' => 'upstream_gene_variant'

The phrasing "in a given gene" is indeed incorrect and should be updated, thank you for pointing that out. It should have been: "Returns the p-value for each SNP around a given gene"


2)
Whereas the documentation says that the stable gene id applies to the location of the eQTL SNP, and no gene information is returned in the result, am I to assume that the stable gene used for the query is also the gene in which the expression-level differences were observed (i.e. no trans effects can be recovered)?  

Correct

Will the result for the same SNP vary according to the query gene? 

Correct

That is, if a particular SNP is in the intergenic region between two oppositely oriented genes, it seems possible for that SNP to have a cis effect on both those genes.  To find such an occurrence, I would have to query by each gene separately?

Correct (or query the SNP)


Thanks for any guidance.

Best regards,

Andrew

_______________________________________________
Dev mailing list    Dev at ensembl.org
Posting guidelines and subscribe/unsubscribe info: http://lists.ensembl.org/mailman/listinfo/dev
Ensembl Blog: http://www.ensembl.info/