genoAssoc: Association analysis with GenABEL

In frankRuehle/systemsbio: Streamlined Analysis and Integration of Systems Biology Data

Description

genoAssoc takes genotype data in GenABEL gwaa format and performs association analysis.

Usage

genoAssoc(
  gwaa,
  projectfolder = "GT/Assoc_GenAbel",
  projectname = NULL,
  trait.name,
  trait.type = "binomial",
  fun.assoc = "mlreg",
  gtmode = "additive",
  times = 1,
  quiet = FALSE,
  TopHitsReported = 1000
)

Arguments

gwaa	gwaa object from GenABEL
projectfolder	character containing path to output folder (will be generated if not existing).
projectname	character used as suffix for output files.
trait.name	character indicating trait(s) of interest.
trait.type	character with data type of analysed trait. Either "gaussian", "binomial" or "guess" (latter option guesses trait type).
fun.assoc	character with association function from GenABEL package to be used. Either mlreg for linear or logistic regression or qtscore for fast score test.
gtmode	character with mode of inheritance if regression is used. Either additive, dominant, recessive or overdominant.
times	If more than one, the number of permutations to be used for calculationg empirical p-values. Relevant for score test only.
quiet	logical Do not print warning messages.
TopHitsReported	numeric with number of top hits to report. Use nsnps(gwaa.object) for all SNPs.

Details

GenABEL package is used for association analysis of a gwaa object. The trait of interest is given in trait.name either as single trait (e.g. affection) or as formula (e.g. affection01 ~ Age + sex) if association shall be adjusted for covariates. All traitnames in trait.name must exist within the phenotype data of the gwaa object. Use phdata(gwaa) for checking which trait names are avalable. Either linear or logistic regression or fast score test can be used for association as given in fun.assoc. In case of regression, the mode of inheritance is given in gtmode. If the fast score test from GenABELs qtscore-function is used without covariates, the function is equivalent to Armitage TREND Test. If covariates are used, the trait is analysed using GLM. Set times to more than 1, to calculate empirical p-values for the score test using times permutations. A table and a Manhattan plot with association results are stored in projectfolder.

Value

scan.gwaa object with Association results. Intermediary results and plots are stored in projectfolder as side effects.

Author(s)

Frank Ruehle

frankRuehle/systemsbio documentation built on Sept. 14, 2020, 1:18 a.m.