On Fri, Jan 29, 2010 at 11:59 AM, Dan Staines <[email protected]> wrote:
> Hi folks, > > A quick question for the list.. We're currently trying to assess how marts > for ensembl variation mart will meet some requirements from one of our > collaborators, so would be interested in how we might achieve some of them > (irrespective of the current mart implementation e.g. links etc.) > > One of the use cases involves selecting a group of genes (based on > phenotype) and then trying to find snps that lie within a specified sequence > window upstream of these genes. As far as I can see, both of these steps are > easy to do with martview but can't see a way of selecting information from > the results of one query to feed into another query without having to carry > out some kind of scripting around a mart API call/sql query. We have some > vague memories of query chaining being mentioned in the past (its alluded to > in some online presentations etc.) but don't see anything beyond the usual > dataset links. Any comments on how this might be achieved would be very > welcome, either with the current or the future implementation of mart. > > So the SNP query you want is something like: http://www.ensembl.org/biomart/martview?VIRTUALSCHEMANAME=default&ATTRIBUTES=hsapiens_gene_ensembl.default.snp.ensembl_gene_id|hsapiens_gene_ensembl.default.snp.external_id|hsapiens_gene_ensembl.default.snp.synonymous_status|hsapiens_gene_ensembl.default.snp.snp_chromosome_strand|hsapiens_gene_ensembl.default.snp.chromosome_location&FILTERS=hsapiens_gene_ensembl.default.filters.ensembl_gene_id."ENSG00000212912,ENSG00000000003,ENSG00000000005,ENSG00000000419,ENSG00000000457,ENSG00000000460,ENSG00000000938,ENSG00000000971,ENSG00000001036,ENSG00000001084,ENSG00000001167"|hsapiens_gene_ensembl.default.filters.snptype_filters."UPSTREAM"&VISIBLEPANEL=resultspanel But instead of pasting in the gene IDs you want this to be the result of another BioMart query, correct? Is the gene query one using the same ensembl genes dataset - if so you can just set the appropriate filters along with upstream snps only? If its a novel dataset of yours then you just need to set up dataset linking between human ensembl genes and your dataset on ensembl gene IDs. Or maybe I'm missing something? Cheers Damian Many thanks, > > Dan. > > -- > Dan Staines, PhD Ensembl Genomes Technical Coordinator > EMBL-EBI Tel: +44-(0)1223-492507 > Wellcome Trust Genome Campus Fax: +44-(0)1223-494468 > Cambridge CB10 1SD, UK http://www.ensemblgenomes.org/ >
