mia: Multiple imputation analysis for hap

View source: R/mia.R

miaR Documentation

Multiple imputation analysis for hap

Description

Multiple imputation analysis for hap

Usage

mia(
  hapfile = "hap.out",
  assfile = "assign.out",
  miafile = "mia.out",
  so = 0,
  ns = 0,
  mi = 0,
  allsnps = 0,
  sas = 0
)

Arguments

hapfile

hap haplotype output file name.

assfile

hap assignment output file name.

miafile

mia output file name.

so

to generate results according to subject order.

ns

do not sort in subject order.

mi

number of multiple imputations used in hap.

allsnps

all loci are SNPs.

sas

produce SAS data step program.

Details

This command reads outputs from hap session that uses multiple imputations, i.e. -mi# option. To simplify matters it assumes -ss option is specified together with -mi option there.

This is a very naive version of MIANALYZE, but can produce results for PROC MIANALYZE of SAS.

It simply extracts outputs from hap.

Value

The returned value is a list.

Note

adapted from hap, in fact cline.c and cline.h are not used. keywords utilities

References

Zhao JH and W Qian (2003) Association analysis of unrelated individuals using polymorphic genetic markers. RSS 2003, Hassalt, Belgium

Clayton DG (2001) SNPHAP. https://github.com/chr1swallace/snphap.

See Also

hap

Examples

## Not run: 
# 4 SNP example, to generate hap.out and assign.out alone
data(fsnps)
hap(id=fsnps[,1],data=fsnps[,3:10],nloci=4)

# to generate results of imputations
control <- hap.control(ss=1,mi=5)
hap(id=fsnps[,1],data=fsnps[,3:10],nloci=4,control=control)

# to extract information from the second run above
mia(so=1,ns=1,mi=5)
file.show("mia.out")

## commands to check out where the output files are as follows:
## Windows
# system("command.com")
## Unix
# system("csh")

## End(Not run)


gap documentation built on Aug. 26, 2023, 5:07 p.m.