Retrieving All Available Frequency Data For A Snp Using Ensembl Api Tools

Question

[Krisr]

Hi,

I would like to retrieve all available SNP population frequency data, for ~5000 SNPs. I have used the API perl variation tools to access the ENsembl data for transcript features, however, I can't figure out how to retrieve frequency information for the SNP, there is little documentation in the API variation tutorial on this. Is anyone familiar with what routines can provide this info, or if there are any to do such a thing?

So, if I had a SNP rsXXXXXX, I would like to retrieve all population frequency data available, HapMap population, CEPH, etc.

Thanks!!

[Krisr]

Thanks, I will try this!

I realize I'm supposed to post responses in the comments, but this wouldn't fit :)

I was trying to piece together some code from the various API Tutorials.

I have the following, but am not sure where, or how, to call the population 'fetch_all_hapmap_populations'. I want to call this on the rs# to retrieve the data, but how do I link the variation object to the population object?

#!/usr/bin/perl -w
use strict;
use warnings;
use Bio::EnsEMBL::Registry;

my $registry = 'Bio::EnsEMBL::Registry';

$registry->load_registry_from_db(
    -host => 'ensembldb.ensembl.org',
    -user => 'anonymous'
);

my $va_adaptor = $registry->get_adaptor('human', 'variation', 'variation'); #get the different adaptors for the different objects needed
my $vf_adaptor = $registry->get_adaptor('human', 'variation', 'variationfeature');
my $pa = $registry->get_adaptor("human","variation","population");

my $pop = $pa->fetch_all_HapMap_Populations();

my $variation = $va_adaptor->fetch_by_name('rs123');

 if (defined $variation)
{
    foreach my $allele (@{$variation->get_all_Alleles()}) 
    {
       print $pop->$allele->fetch_all_HapMap_Populations() . "\t" . $allele->allele . "\t" . $allele->frequency . "\n";
    }
}

[Krisr]

I applied the above MySQL to a DBI interface with perl. I've made a loop to iterate through all SNPs and report back the freqs. Thanks again!!!

Answer 1

[Pierre Lindenbaum]

I'm not an expert on the mysql schema for ensembl, so use my answer with caution; The following SQL query seems to return the data you want:

mysql -u anonymous -h ensembldb.ensembl.org -P 5306 -D homo_sapiens_core_61_37f -A
  select distinct 
V.name,
S.handle,
A.allele,
A.frequency,
M.name,
F.allele_string
  from (
allele as A,
variation as V,
subsnp_handle as S,
variation_feature as F
  ) left join
  sample as M
on (M.sample_id = A.sample_id
  ) where 
V.variation_id=A.variation_id and
S.subsnp_id =A.subsnp_id and
F.variation_id=V.variation_id and 
V.name="rs3"
  order by 1,2,5;

results:

	+------+--------------+--------+-----------+---------------------------------+---------------+
	| name | handle | allele | frequency | name | allele_string |
	+------+--------------+--------+-----------+---------------------------------+---------------+
	| rs3 | 1000GENOMES | C | NULL | NULL | C/T |
	| rs3 | 1000GENOMES | T | NULL | NULL | C/T |
	| rs3 | 1000GENOMES | C | NULL | 1000GENOMES:CEU | C/T |
	| rs3 | 1000GENOMES | T | NULL | 1000GENOMES:CEU | C/T |
	| rs3 | 1000GENOMES | C | 0.958333 | 1000GENOMES:low_coverage:CEU | C/T |
	| rs3 | 1000GENOMES | T | 0.0416667 | 1000GENOMES:low_coverage:CEU | C/T |
	| rs3 | 1000GENOMES | C | 0.958333 | 1000GENOMES:low_coverage:CHBJPT | C/T |
	| rs3 | 1000GENOMES | T | 0.0416667 | 1000GENOMES:low_coverage:CHBJPT | C/T |
	| rs3 | 1000GENOMES | C | 0.838983 | 1000GENOMES:low_coverage:YRI | C/T |
	| rs3 | 1000GENOMES | T | 0.161017 | 1000GENOMES:low_coverage:YRI | C/T |
	| rs3 | 1000GENOMES | C | 0.958333 | 1000GENOMES:pilot.1.CEU | C/T |
	| rs3 | 1000GENOMES | T | 0.0416667 | 1000GENOMES:pilot.1.CEU | C/T |
	| rs3 | 1000GENOMES | C | 0.84 | 1000GENOMES:pilot.1.YRI | C/T |
	| rs3 | 1000GENOMES | T | 0.16 | 1000GENOMES:pilot.1.YRI | C/T |
	| rs3 | 1000GENOMES | C | NULL | 1000GENOMES:YRI | C/T |
	| rs3 | 1000GENOMES | T | NULL | 1000GENOMES:YRI | C/T |
	| rs3 | BCM_SSAHASNP | C | NULL | NULL | C/T |
	| rs3 | BCM_SSAHASNP | T | NULL | NULL | C/T |
	| rs3 | KWOK | C | NULL | NULL | C/T |
	| rs3 | KWOK | T | NULL | NULL | C/T |
	| rs3 | KWOK | C | 0.967 | CSHL-HAPMAP:HapMap-CEU | C/T |
	| rs3 | KWOK | T | 0.033 | CSHL-HAPMAP:HapMap-CEU | C/T |
	| rs3 | KWOK | C | 0.956 | CSHL-HAPMAP:HapMap-HCB | C/T |
	| rs3 | KWOK | T | 0.044 | CSHL-HAPMAP:HapMap-HCB | C/T |
	| rs3 | KWOK | C | 0.978 | CSHL-HAPMAP:HapMap-JPT | C/T |
	| rs3 | KWOK | T | 0.022 | CSHL-HAPMAP:HapMap-JPT | C/T |
	| rs3 | KWOK | C | 0.89 | CSHL-HAPMAP:HapMap-YRI | C/T |
	| rs3 | KWOK | T | 0.11 | CSHL-HAPMAP:HapMap-YRI | C/T |
	| rs3 | KWOK | C | 0.95 | KWOK:B | C/T |
	| rs3 | KWOK | T | 0.05 | KWOK:B | C/T |
	| rs3 | KWOK | C | 0.96 | KWOK:C | C/T |
	| rs3 | KWOK | T | 0.04 | KWOK:C | C/T |
	| rs3 | KWOK | C | 0.89 | KWOK:H | C/T |
	| rs3 | KWOK | T | 0.11 | KWOK:H | C/T |
	| rs3 | KWOK | C | 0.93 | KWOK:S | C/T |
	| rs3 | KWOK | T | 0.07 | KWOK:S | C/T |
	| rs3 | PERLEGEN | C | 0.673913 | PERLEGEN:AFD_AFR_PANEL | C/T |
	| rs3 | PERLEGEN | T | 0.326087 | PERLEGEN:AFD_AFR_PANEL | C/T |
	| rs3 | PERLEGEN | C | 0.958333 | PERLEGEN:AFD_CHN_PANEL | C/T |
	| rs3 | PERLEGEN | T | 0.0416667 | PERLEGEN:AFD_CHN_PANEL | C/T |
	| rs3 | PERLEGEN | C | 0.979167 | PERLEGEN:AFD_EUR_PANEL | C/T |
	| rs3 | PERLEGEN | T | 0.0208333 | PERLEGEN:AFD_EUR_PANEL | C/T |
	| rs3 | SC_SNP | C | NULL | NULL | C/T |
	| rs3 | SC_SNP | T | NULL | NULL | C/T |
	+------+--------------+--------+-----------+---------------------------------+---------------+
	44 rows in set (0.09 sec)

view raw biostars-6703.txt hosted with ❤ by GitHub

[Giovanni M Dall'Olio]

ERROR 1146 (42S02): Table 'homo_sapiens_core_61_37f.allele' doesn't exist

[Giovanni M Dall'Olio]

The correct database name is homo_sapiens_variation_61_37 , instead of homo_sapiens_core_61_37f

[Pierre Lindenbaum]

Thanks Giovanni

[Khader Shameer]

I think the correct table is homo_sapiens_variation_61_37f, but am not sure why the table retrieving different values when I compared to 1000 genomes MAFs in dbSNP.

Answer 2

[Bert Overduin]

Hi,

This script should do the trick:

#!/usr/local/bin/perl
use strict;
use warnings;

use Bio::EnsEMBL::Registry;

my $reg = 'Bio::EnsEMBL::Registry';

$reg->load_registry_from_db(
    -host => 'ensembldb.ensembl.org',
    -user => 'anonymous'
);

# get a variation adaptor
my $va = $reg->get_adaptor("human", "variation", "variation");

# fetch the variation by name
my $v = $va->fetch_by_name('rs123');

# get all the alleles
my @alleles = @{$v->get_all_Alleles()};

foreach my $allele(@alleles) {
    print
        $v->name, "\t",
        $allele->allele, "\t",
        (defined($allele->frequency) ? $allele->frequency : "-"), "\t",
        (defined($allele->population) ? $allele->population->name : "-"), "\n";
}

Hope this helps.

Note that questions about Ensembl and the Ensembl API can also be posted to the Ensembl helpdesk or the Ensembl developers mailing list (how to subscribe to this list you can find here).