CSS: Composite Selection Signal
In NESCent/MINOTAUR: MultIvariate visualisatioN and OuTlier Analysis Using R

Description Usage Arguments Details Author(s) References

Calculates the CSS statistic for each row (locus, SNP) in the data frame. Data are subset prior to calculating distances (see details).

1
2
3

CSS(dfv, column.nums = 1:ncol(dfv), subset = 1:nrow(dfv),
  two.tailed = rep(TRUE, length(column.nums)), right.tailed = rep(FALSE,
  length(column.nums)))

`dfv`	a data frame containing observations in rows and statistics in columns.
`column.nums`	indexes the columns of the data frame that will be used to calculate CSS (all other columns are ignored).
`subset`	index the rows of the data frame that fractional ranks will be relative to.
`two.tailed`	a boolean vector with one entry for each chosen column, where TRUE indicates that the column should be converted to fractinal ranks based on a two-tailed test.
`right.tailed`	a boolean vector with one entry for each chosen column, where TRUE indicates that the column should be converted to fractional ranks based on a right-tailed test (see `?stat_to_pvalue`).

CSS is calculated based on the method described in Randhawa et al (2014). Selected columns of dfv are first converted to fractional ranks (see ?stat_to_pvalue). Fractional ranks are then converted to z-scores using the inverse cumulative normal transformation. The mean z-score is then taken over variables, and converted to a p-value based on the appropriate normal distribution. Finally, the CSS statistic is defined as -log(p-value) in base 10.

As fractional ranks are obtained using the stat_to_pvalue function, the various arguments to this function are available. This includes options for calculating fractional ranks based on one- and two-tailed methods for each variable independently.

Robert Verity r.verity@imperial.ac.uk

Randhawa, Imtiaz Ahmed Sajid, et al. "Composite selection signals can localize the trait specific genomic regions in multi-breed populations of cattle and sheep." BMC genetics 15.1 (2014): 1.

NESCent/MINOTAUR documentation built on May 7, 2019, 6:01 p.m.