optimize_qtl: Model optimization
In guilherme-pereira/QTLpoly: Random-effect multiple QTL mapping in autopolyploids

Description Usage Arguments Value Author(s) References See Also Examples

Tests each QTL at a time and updates its position (if it changes) or drops the QTL (if non-significant).

optimize_qtl(
  data,
  offset.data = NULL,
  model,
  sig.bwd = 0.05,
  score.null = NULL,
  polygenes = FALSE,
  n.clusters = NULL,
  plot = "optimize",
  verbose = TRUE
)

## S3 method for class 'qtlpoly.optimize'
print(x, pheno.col = NULL)

`data`	an object of class `qtlpoly.data`.
`offset.data`	a data frame with the same dimensions of `data$pheno` containing offset variables; if `NULL` (default), no offset variables are considered.
`model`	an object of class `qtlpoly.model` containing the QTL to be optimized.
`sig.bwd`	the desired score-based p-value threshold for backward elimination, e.g. 0.0001 (default).
`score.null`	an object of class `qtlpoly.null` with results of score statistics from resampling.
`polygenes`	if `TRUE` all QTL but the one being tested are treated as a single polygenic effect, if `FALSE` (default) all QTL effect variances have to estimated.
`n.clusters`	number of parallel processes to spawn.
`plot`	a suffix for the file's name containing plots of every QTL optimization round, e.g. "optimize" (default); if `NULL`, no file is produced.
`verbose`	if `TRUE` (default), current progress is shown; if `FALSE`, no output is produced.
`x`	an object of class `qtlpoly.optimize` to be printed.
`pheno.col`	a numeric vector with the phenotype columns to be printed; if `NULL`, all phenotypes from `'data'` will be included.

An object of class qtlpoly.optimize which contains a list of results for each trait with the following components:

`pheno.col`	a phenotype column number.
`stat`	a vector containing values from score statistics.
`pval`	a vector containing p-values from score statistics.
`qtls`	a data frame with information from the mapped QTL.

Guilherme da Silva Pereira, gdasilv@ncsu.edu

Pereira GS, Gemenet DC, Mollinari M, Olukolu BA, Wood JC, Mosquera V, Gruneberg WJ, Khan A, Buell CR, Yencho GC, Zeng ZB (2020) Multiple QTL mapping in autopolyploids: a random-effect model approach with application in a hexaploid sweetpotato full-sib population, Genetics 215 (3): 579-595. http://doi.org/10.1534/genetics.120.303080.

Qu L, Guennel T, Marshall SL (2013) Linear score tests for variance components in linear mixed models and applications to genetic association studies. Biometrics 69 (4): 883–92. doi.org/10.1111/biom.12095.

Zou F, Fine JP, Hu J, Lin DY (2004) An efficient resampling method for assessing genome-wide statistical significance in mapping quantitative trait loci. Genetics 168 (4): 2307-16. doi.org/10.1534/genetics.104.031427

read_data, null_model, search_qtl

  ## Not run: 
  # load raw data
  data(maps)
  data(pheno)

  # estimate conditional probabilities using 'mappoly' package
  library(mappoly)
  genoprob <- lapply(maps, calc_genoprob)

  # prepare data
  data <- read_data(ploidy = 6, geno.prob = genoprob, pheno = pheno, step = 1)

  # build null models
  null.mod <- null_model(data = data, n.clusters = 4, plot = "null")

  # perform forward search
  search.mod <- search_qtl(data = data, model = null.mod, w.size = 15, sig.fwd = 0.01,
    n.clusters = 4, plot = "search")

  # optimize model
  optimize.mod <- optimize_qtl(data = data, model = search.mod, sig.bwd = 0.0001,
    n.clusters = 4, plot = "optimize")
  
## End(Not run)