The GWASapi package provides access to the NHGRI-EBI catalog of GWAS summary statistics. For details on the API, see its documentation, as well as Pjotr Prins’s documentation at github.
You can install GWASapi from GitHub.
You first need to install the devtools.
Then use devtools::install_github() to install
GWASapi.
Load the package with library().
The purpose of the GWASapi package is to provide access to summary statistics for human GWAS. First, you can get lists of studies and traits that are available.
To get lists of studies, use list_studies(). The default
is to return just 20 studies. You can control that limit with the
argument size. You can also use start to step
through the full set.
## [1] "GCST000028" "GCST000392" "GCST000510" "GCST000553" "GCST000568"To retrieve all studies, set a higher limit
## [1] 1051To get a list of traits, use list_traits(). Again the
default is to return just 20 values. To get all traits, use the
size argument.
## [1] 505The traits are returned as identifiers like EFO_0000249. To get a description of a trait, you can use the ontology lookup service, for example https://www.ebi.ac.uk/efo/EFO_0001360
Chromosomes are stored as integers 1-24.
## [1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24To get associations for a specific variant by its rs-number, use
get_variant(). If you know the chromosome it is on, you’ll
get faster results by providing the chromosome. And again, the default
is to return just 20 values, so use the size and
start arguments if you want a comprehensive list.
## p_value study_accession trait
## 0 0.12280000 GCST000028 EFO_0001360
## 1 0.02980604 GCST000392 EFO_0001359
## 2 0.26000000 GCST000510 EFO_0004309
## 3 0.63000000 GCST000553 EFO_0004839
## 4 0.39340000 GCST000568 EFO_0004465Use the arguments p_lower and p_upper to
focus on associations with p-value in a specified range. For example, to
get all of the associations with p-value < 10-10, you
would do:
result <- get_variant("rs2228603", 19, p_upper=1e-10)
result[,c("p_value", "study_accession", "trait")]## p_value study_accession trait
## 0 1.739000e-57 GCST002216 EFO_0004530
## 1 1.049000e-62 GCST002221 EFO_0004574
## 2 4.433000e-44 GCST002222 EFO_0004611
## 3 3.652000e-11 GCST007516 EFO_0001360
## 4 3.652000e-11 GCST007518 EFO_0001360
## 5 2.225074e-308 GCST010144 EFO_0004631
## 6 4.782919e-13 GCST010772 EFO_0004918
## 7 2.926260e-17 GCST90000614 EFO_0004631
## 8 6.308030e-18 GCST90000615 EFO_0004631
## 9 3.616430e-17 GCST90000616 EFO_0004631
## 10 1.431780e-15 GCST90000617 EFO_0004631
## 11 2.721982e-14 GCST90000618 EFO_0004631
## 12 1.800000e-205 GCST90002412 EFO_0004611
## 13 3.750000e-64 GCST90013663 EFO_0004735
## 14 1.535000e-33 GCST90013664 EFO_0004736
## 15 8.800000e-65 GCST90016673 EFO_0010821To get associations for a specific region, use
get_asso(). For example, to get the region from 19.2 Mbp to
19.3 Mbp on chr 19:
result <- get_asso(chr=19, bp_lower=19200000, bp_upper=19300000)
result[,c("chromosome", "base_pair_location", "p_value", "study_accession", "trait")]## chromosome base_pair_location p_value study_accession trait
## 0 19 19219115 0.12280000 GCST000028 EFO_0001360
## 1 19 19230637 0.12400000 GCST000028 EFO_0001360
## 2 19 19233770 0.12400000 GCST000028 EFO_0001360
## 3 19 19238138 0.27990000 GCST000028 EFO_0001360
## 4 19 19250926 0.48100000 GCST000028 EFO_0001360
## 5 19 19262880 0.76410000 GCST000028 EFO_0001360
## 6 19 19263252 0.48500000 GCST000028 EFO_0001360
## 7 19 19296909 0.13780000 GCST000028 EFO_0001360
## 8 19 19208675 0.34851112 GCST000392 EFO_0001359
## 9 19 19219115 0.02980604 GCST000392 EFO_0001359
## 10 19 19224274 0.72647403 GCST000392 EFO_0001359
## 11 19 19225799 0.09019226 GCST000392 EFO_0001359
## 12 19 19230637 0.59469796 GCST000392 EFO_0001359
## 13 19 19233770 0.93955453 GCST000392 EFO_0001359
## 14 19 19247863 0.76167551 GCST000392 EFO_0001359
## 15 19 19250926 0.24164792 GCST000392 EFO_0001359
## 16 19 19274602 0.71375546 GCST000392 EFO_0001359
## 17 19 19277637 0.43687139 GCST000392 EFO_0001359
## 18 19 19280236 0.76699603 GCST000392 EFO_0001359
## 19 19 19281592 0.70582591 GCST000392 EFO_0001359You can restrict those results to a particular study.
result <- get_asso(chr=19, bp_lower=19200000, bp_upper=19300000, study="GCST000392")
result[,c("chromosome", "base_pair_location", "p_value", "study_accession", "trait")]## chromosome base_pair_location p_value study_accession trait
## 0 19 19208675 0.34851112 GCST000392 EFO_0001359
## 1 19 19219115 0.02980604 GCST000392 EFO_0001359
## 2 19 19224274 0.72647403 GCST000392 EFO_0001359
## 3 19 19225799 0.09019226 GCST000392 EFO_0001359
## 4 19 19230637 0.59469796 GCST000392 EFO_0001359
## 5 19 19233770 0.93955453 GCST000392 EFO_0001359
## 6 19 19247863 0.76167551 GCST000392 EFO_0001359
## 7 19 19250926 0.24164792 GCST000392 EFO_0001359
## 8 19 19274602 0.71375546 GCST000392 EFO_0001359
## 9 19 19277637 0.43687139 GCST000392 EFO_0001359
## 10 19 19280236 0.76699603 GCST000392 EFO_0001359
## 11 19 19281592 0.70582591 GCST000392 EFO_0001359
## 12 19 19297355 0.78331140 GCST000392 EFO_0001359To get associations for a given trait, use
get_trait_asso(). You can’t restrict this to a given
chromosome region.
## [1] 71## chromosome base_pair_location p_value study_accession trait
## 0 14 63485893 0.00e+00 GCST006801 EFO_0001360
## 1 8 141795579 0.00e+00 GCST006801 EFO_0001360
## 2 10 112988738 4.38e-109 GCST006867 EFO_0001360
## 3 10 112988858 4.96e-127 GCST006867 EFO_0001360
## 4 10 112989975 2.98e-221 GCST006867 EFO_0001360