pop_predict_quick: Quicker procedures for PopVar

Description Usage Arguments Details Examples

View source: R/pop_predict_quick.R

Description

Calculates the expected mean, genetic variance, and superior progeny mean of a bi-parental population based on genomewide marker effects.

Usage

1
2
3
pop_predict_quick(G.in, y.in, map.in, crossing.table, parents, tail.p = 0.1,
  model = c("RRBLUP"), map.function = c("haldane", "kosambi", "cf",
  "morgan"))

Arguments

y.in

A data.frame of entry names and phenotypic value. The first column must be the entry name, and subsequent columns must contain phenotypic values. Column names must be the names of the traits.

map.in

A data.frame of genetic map information for the markers in G.in. The same number of markers must be present in map.in as in G.in. The first column must be the marker name, the second column must be the chromosome, and the third column must be the genetic map position (in cM).

crossing.table

A data.frame of parental combinations from which to simulate biparental population. The first column must be parent 1, and the second column must be parent 2. All entries in crossing.table must have genotypic information in G.in.

parents

If crossing.table is not passed, a vector of entry names to serve as parents must be passed. All possible combinations of n parents (i.e. choose(n, 2)) will be simulated.

tail.p

The proportion of the distribution of genotypic values from which to define the superior progeny mean. Defaults to 0.1.

model

The statistical model from which to predict marker effects.

map.function

The map function to use when converting genetic map distance to recombination rate.

pheno

A data.frame of phenotypic data. The first column must contain line/genotype, and subsequent columns may contain factors to be modeled as fixed (e.g. environment), or trait values.

geno

A data.frame of marker names, positions, and genotypes. The first column must contain marker names, the second column chromosome, the third column position, and remaining columns must be the genotype calls for each line/genotype at the respective marker. Must be coded at z = {-1, 0, 1}, where 1 is homozygous for the first allele, -1 is homozygous for the second allele, and 0 is heterozygous.

Details

This functions executes the same predictions as the function pop.predict, but uses the expectation of the genetic variance in a biparental population, as opposed to simulating biparental populations. The approach using the expectation is about 250 times faster than using simulated biparental population on the default settings.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
# Load example data
library(dplyr)
data("genos")
data("phenos")
data("map")

# Create a crossing table
crossing.table <- combn(x = row.names(genos), m = 2) %>%
  t() %>%
  as.data.frame() %>%
  structure(names = c("parent1", "parent2")) %>%
  sample_n(100)


pp_quick_out <- pop_predict_quick(G.in = genos, y.in = phenos, map.in = map,
                                  crossing.table = crossing.table)

neyhartj/gws documentation built on Nov. 9, 2017, 8:35 p.m.