This is a test version of Biostars. For the public version, visit https://www.biostars.org.
SNP rs ID batch query

Hi everybody:

I have a list of 4000 SNP rs identifiers from dbSNP and I would like to get their orientation (forward/reverse). However, I have not been able to find the way to do it in a straightforward way, as dbSNP no longer supports batch query apparently.

Any help would be really appreciated.

snp
• 8,664 views • link
updated by Emily Emily • written by lorenzolf90 • lorenzolf90

Aren't all dbSNP variant given on the forward (+) strand of the reference genome regardless of a gene/transcript?

• 1 views • link
written by finswimmer finswimmer

No - some percentage of them are given on the reverse (-), not entirely sure why this is. For example, rs499479

• 1 views • link
written by SpacemanSpiffo • SpacemanSpiffo

I realized by chance that the bed files for dbSNP contain the strand information.

So one way you can choose, is to first find out on which chromosome your SNPs are located and query than the corresponding bed file for it.

fin swimmer

• 1 views • link
written by finswimmer finswimmer

1 answer

Care Lorenzo / Dear Lorenzo,

You can do this in varying ways, one being via biomaRt in R:

require(biomaRt)

ensembl <- useMart("ENSEMBL_MART_SNP", dataset = "hsapiens_snp")

getBM(attributes=c(
    "refsnp_id", "chr_name", "chrom_start", "chrom_end", "chrom_strand",
    "allele", "mapweight", "validated", "allele_1", "minor_allele",
    "minor_allele_freq", "minor_allele_count", "clinical_significance",
    "synonym_name", "ensembl_gene_stable_id"),
    filters="snp_filter", values="rs6025",
    mart=ensembl, uniqueRows=TRUE)

refsnp_id chr_name chrom_start  chrom_end   chrom_strand allele mapweight
rs6025    1        169549811    169549811   1            C/T    1

validated
1000Genomes,Cited,ESP,ExAC,Frequency,gnomAD,HapMap,Phenotype_or_Disease,TOPMed

allele_1 minor_allele minor_allele_freq minor_allele_count
C        TRUE         0.00599042        30

clinical_significance
benign,pathogenic,drug response,risk factor

synonym_name ensembl_gene_stable_id
17284        ENSG00000198734

The value of strand will be +1 (plus strand) or -1 (minus strand).

------------------------------------

To look up multiple records at the same time, pass a vector of rs IDs to getBM() as the values parameter, for example:

snps <- c("rs1", "rs2", "rs3", ..., "rsx")

getBM(attributes=c(
        "refsnp_id", "chr_name", ...),
        filters="snp_filter", values=snps,
        mart=ensembl, uniqueRows=TRUE)

Regarding strand orientation in dbSNP, you may read the FAQ, starting from: Strand Orientation.

Kevin

• 1 views • link
written by Kevin Blighe Kevin Blighe

Dear Kevin, thank you so much for your through answer, it was really useful. However, I am afraid it does not work as I expected since when I use biomaRt with rs36563, which I know for certain is in REVERSE (https://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=36563) I still get a +1 value in the strand field:

*require(biomaRt)
ensembl <- useMart("ENSEMBL_MART_SNP", dataset = "hsapiens_snp")
getBM(attributes=c(
    "refsnp_id", "chr_name", "chrom_start", "chrom_end", "chrom_strand",
    "allele", "mapweight", "validated", "allele_1", "minor_allele",
    "minor_allele_freq", "minor_allele_count", "clinical_significance",
    "synonym_name", "ensembl_gene_stable_id"),
    filters="snp_filter", values="rs36563",
    mart=ensembl, uniqueRows=TRUE)

refsnp_id chr_name chrom_start  chrom_end   chrom_strand allele mapweight

rs36563    14        70885931    70885931   1            T/G    1*

This is the same value that I obtain with FORWARD SNPs (such as rs6902771), so I still cannot differentiate them. Any ideas how I can do so?

Thank you so much again!

• 1 views • link
updated by finswimmer finswimmer • written by lorenzolf90 • lorenzolf90

Hello,

could you please explain why it is important to know whether a variant is reported on the reverse or forward strand in dbSNP?

fin swimmer

• 1 views • link
written by finswimmer finswimmer

Hello Fin,

Coming back to the rs36563 example, the reported risk allele for a particular trait is A (https://www.ebi.ac.uk/gwas/search?query=rs36563), however, when I annotate such SNP using biomaRt (as I put above) I find that the alleles are T/G, and that is because the SNP is on the reverse strand.

That is why I would like to know the strand of a batch o variants, so I can consider the complementary nucleotide in those cases where the SNP is on the reverse and therefore see if it matches with the risk allele.

Hope that makes it clearer.

Thank you very much!

• 1 views • link
written by lorenzolf90 • lorenzolf90

Variants do not have strand, which is why Ensembl adopt a default behaviour of always showing the forward strand alleles for all variants. The risk allele reported in papers is entirely dependant on what the paper authors decide to do, which may be the forward strand, may follow the strand of the alleles quoted by dbSNP, or may not.

• 1 views • link
written by Emily Emily

Hi, I tried this method and I'm getting the following error message, can you please suggest how I can address this issue? Thanks in advnace

Error in curl::curl_fetch_memory(url, handle = handle) : Timeout was reached: [www.ensembl.org:443] Operation timed out after 300004 milliseconds with 309191 bytes received

• 0 views • link
written by Mahan Mahan

Are you overloading Ensembl's server?

• 0 views • link
written by Kevin Blighe Kevin Blighe

Please create a new question with full information of your BioMart query

• 0 views • link
written by Emily Emily

Log in to answer this question.

More posts like this