Functions

gn_url()

Return the default GeneNetwork server API URL.

All user-level functions optionally take a GeneNetwork server URL that could be different from the standard URL in case one wanted to query a different instance of the server.

Example

julia> gn_url()
"https://genenetwork.org/api/v_pre1/"

check_gn(url::AbstractString)

Check if GeneNetwork server is responding properly.

Returns the HTTP status code (200 if successful) and prints a message.

Example

julia> check_gn()
GeneNetwork is alive.
200

list_datasets(group::String="";gn_url::String=gn_url())

Lists all datasets available in a specified group.

list_groups(species::String="";gn_url::String=gn_url())

If species is not specified, then it returns all groups (segregating populations) represented in the GeneNetwork database. If the string species is specified, then it returns all groups for that species.

list_species(species::String="";gn_url::String=gn_url())

Returns a data frame with a list of species respresented in the GeneNetwork database. If the species string is non-empty, it will return the information of the matched species.

list_geno(group::String; gn_url::String=gn_url())

Returns a data frame with the location name of the different geno files of a group, and if available some metadata such as strain of the first filial generation, maternal and paternal strain. If there exist more than one location, the default location is indicated by a *.

get_geno(group::String,format::String="geno";gn_url::String=gn_url())

Returns the genotype matrix for a group in a given format.

Currently works only for files in the geno format.

get_pheno(dataset::String;gn_url::String=gn_url())

Returns the non-omic ("clinical") phenotypes for a given dataset.

get_pheno(dataset::String,trait::String; gn_url::String=gn_url())

Returns a given trait in a group.

get_omics(dataset::String;gn_url::String=gn_url())

Returns the omic phenotypes for a given dataset.

info_dataset(dataset::String, trait::String=""; gn_url::String = gn_url())

Returns information about a dataset (if trait is empty), or else a given trait in a specified dataset.

info_pheno(group::String; gn_url::String=gn_url())

Returns the maximum LRS for each trait in a group`.

info_pheno(group::String, trait::String; gn_url::String=gn_url())

Returns the maximum LRS for each trait in a group, or alternatively a probe (use trait) in dataset (use group).

download_geno(group::String,format::String="geno";
                gn_url::String=gn_url(),
                path::String = "")

Download the genotype matrix for a group in a given format.

Currently works only for files in the geno format.

The file will be downloaded in path location, if path is not specified, then the file will be a temporary file.

download_pheno(dataset::String;gn_url::String=gn_url())

Downloads the non-omic ("clinical") phenotypes for a given dataset. The file will be downloaded in path location, if path is not specified, then the file will be a temporary file.

download_omics(dataset::String; 
                gn_url::String=gn_url(), 
                path::String = tempname())

Download the omic phenotypes for a given dataset. The file will be downloaded in path location, if path is not specified, then the file will be a temporary file.

show_table(df::DataFrame)

Displays a dataframe in pretty mode.

makerectangular(varrays::Vector{Vector{T}} where T)

Takes an array of arrays of any type and make it a matrix of String. Arrays can be of different type and length. To make the data rectangular, it uses the filler argument, by default filler is an empty string.

Example 1

julia> myArrayOfArrays = [["A1","A2", "A3"], ["B1", "B2"], ["C1","C2"]];
julia> make_rectangular(test)
3×3 Matrix{String}:
"A1"  "B1"  "C1"
"A2"  "B2"  "C2"
"A3"  ""    ""

Example 2

julia> myArrayOfArrays = [["A1","A2", "A3"], [1, 2], ["C1","C2"]];
julia> make_rectangular(test)
3×3 Matrix{String}:
"A1"  "1"  "C1"
"A2"  "2"  "C2"
"A3"  ""    ""

has_error_500(s::String)

Returns true if the server encountered an internal error and was unable to complete the request, otherwise it returns false.

parse_geno(filename::String)

Function to parse a .geno file, and returns the genotype dataframe.

genofile_location(json_parsed::Dict)

Returns a vector of String containing the location of geno files for a group.

genofile_location(group::String)

Returns a vector of String containing the location of geno files for a group.

has_genofile_meta(group::String; gn_url::String=gn_url())

Returns true if the group's genotype files possesse metadata that may contain alternative location of the files.

run_correlation(trait::String, dataset::String, group::String; tp:;String="sample",
                method::String="pearson", n_result::Int64=500,
                gn_url::String=gn_url())

This function correlates a trait in a dataset against all traits in a target database (group).

This query currently takes the following parameters:

• trait (required): trait ID for which the correlation is wanted
• dataset (required): dataset in which the trait occurs (use the short abbreviation)
• group (required): group in which which the comparisons will be performed
• tp: sample (default) | tissue
• method: pearson (default) | spearman
• n_result: maximum number of results to return (default = 500)

run_gemma(dataset::String,trait::String; use_loco::Bool=false, maf::Float64=0.01,
    gn_url::String=gn_url())

Runs GEMMA with the specified trait in the specified dataset. Optional arguments include:

• use_loco (default=false): if LOCO (leave one chromosome out) should be used
• maf (default=0.01): the minimum minor allele frequency of the markers to be scanned

function run_rqtl(dataset::String, trait::String; method::String="hk",
              model::String="normal", n_perm::Int64=0,
              control_marker::String="",
              gn_url::String=gn_url())

Runs R/qtl's scanone function with the specified trait in the specified dataset.

Some optional arguments have analogs in options in the parent scanone function. See the help for that function.

Optional arguments inlcude:

• method (default="hk"): possible values are
  • hk: Haley-Knott method
  • em: EM algorithm
  • ehk: Extended EM algorithm
  • imp: multiple imputation
  • mr: marker regression dropping individuals with missing genotypes
  • mr-imp: marker regression filling in missing data with a single imputation
  • mr-argmax: marker regression filling in by the Viterbi algorithm
• model (default="normal"): model for trait given genotypes; possible values are
  • normal: equivalent to linear regression with complete data
  • binary: equivalent to logistic regression with complete data
  • 2part: two-part model for trait given genotype
  • np: non-parametric analysis
• n_perm (default=0) number of permutations to perform
• control_marker: name of marker to control for as an additive covariate